Pan Troglodytes Verus), Senegal, West Africa Stephanie Lynn Bogart Iowa State University

Masthead Logo

Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

1-1-2005 Insectivory, termite diversity, and tool use of the Fongoli chimpanzees (Pan troglodytes verus), Senegal, West Africa Stephanie Lynn Bogart Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd

Recommended Citation Bogart, Stephanie Lynn, "Insectivory, termite diversity, and tool use of the Fongoli chimpanzees (Pan troglodytes verus), Senegal, West Africa" (2005). Retrospective Theses and Dissertations. 18863. https://lib.dr.iastate.edu/rtd/18863

This Thesis is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Insectivory, termite diversity, and tool use of the Fongoli chimpanzees (Pan troglodytes

verus), Senegal, West Africa

Stephanie Lynn Bogart

A thesis submitted to the graduate faculty in partial fulfillment of the

requirements for the degree of

MASTER OF ARTS

Major: Anthropology

Program of Study Committee: Jill D. Pruetz, Major Professor Matthew G. Hill Donald Lewis

Iowa State University Ames, Iowa 2005

Graduate College Iowa State University

This is to certify that the Master' s thesis of

Stephanie Lynn Bogart has met the thesis requirements oflowa State University

Signatures have been redacted for privacy lll

This thesis is dedicated to:

Ester Zepkin, Ileen Bogart, and Diane Bogart:

A long line of strong and independent women. IV

TABLE OF CONTENTS

DEDICATION m

LIST OF FIGURES v1

LIST OF TABLES vm

ACKNOWLEDGEMENTS ix

ABSTRACT XI

CHAPTER 1: Introduction 1 Importance of Chimpanzees to Understanding Human Evolution Studies 1 Primate Insectivory 2 Non-Human Ape Diet 4 Chimpanzee insectivory 6 Questions and Goals 7 Study Site and Subjects 8 Implications of This Research 11

CHAPTER 2: Significance of Tool Use 14 What is Tool Use? 14 Animal tool use 16 Primates and Tool Use 17 Ape tool use (Hominoidea) 18 Termite Fishing 21 Implications for Culture 23

CHAPTER 3: Ecology of Study Site 26 Introduction 26 Methods 27 Data collection 28 Analyses 29 Results 29 Discussion and Conclusions 34

CHAPTER 4: Termite Inhabitants 36 Introduction 36 Methods 36 Results 39 Mound characteristics 39 Termite distribution 39 Discussion and Conclusions 41 v

CHAPTER 5: Chimpanzee Termite-fishing Tools 47 Introduction 47 Data Collection 48 Results 48 Plant material and vegetation 48 Tool characteristics 50 Discussion and Conclusions 53

CHAPTER 6: Conclusions and Summary 57 Conclusions of This Study 57 Benefits from Eating Termites 59 Macrotermes 62 Hominid Insectivory? 62 Questions of Culture 65 Future Research 67

REFERENCES CITED 69 Vl

LIST OF FIGURES

Figure 1.1: Senegal, West Africa (Fongoli Chimpanzee Study Site). 10

Figure 1.2: Chimpanzee (Mamadou) of the Fongoli community. 11

Figure 2.1: Three adult males branch clasp grooming at Fongoli, a behavior the 15 exhibits 'cultural' variation across sites. Figure 2.2: Chimpanzee termite fishing. 23

Figure 3.1: Habitats ofFongoli. 28

Figure 3.2: Recorded termite mound distribution with GPS Magellan (UTM units). 30

Figure 3.3: Fongoli habitat distribution. 31

Figure 3.4: Habitat distribution at Fongoli and location of termite mounds used by 31 chimpanzees. Figure 3.5: Termite mound density proportion versus termite fished habitat used by 33 chimpanzees at Fongoli. Figure 4.1: Author and Mbouli Camera (field assistant) collecting termites. 37

Figure 4.2: Mounds at Fongoli. 38

Figure 4.3: Double inhabitant mound (convex with mushroom) in woodland 38 habitat. Figure 4.4: Termite diversity at Fongoli. 40

Figure 4.5: Termite soldiers of the genera associated with tools. 41

Figure 5.1: Chimpanzee holding plant material in mouth. 49

Figure 5.2: Fresh tools found at mound TM161. 50

Figure 5.3: Tool kit of Saba at mound TM161. 50

Figure 5.4: Tool characteristics. 52

Figure 5.5: Perforating stick of bamboo from mound TM152. 52

Figure 5.6: Gombe chimpanzee termite fishing. 56

Figure 6.1: Macrotermes sp. soldier and immatures. 60 Vll

LIST OF FIGURES (Continued)

Figure 6.2: Macrotermes sp. workers and soldiers. 61

Figure 6.3: Adolescent male, West African chimpanzee (Pan troglodytes verus). 67 Vlll

LIST OF TABLES

Table 1.1: Compositions of great ape diet (percentages). 6

Table 3.1: Definitions of habitat types. 28

Table 3.2: Mound density (per hectare) at Fongoli for each category of habitat. 33

Table 4.1: Termite genera and species at chimpanzee sites across Africa. 43

Table 5.1: Flexible probe measurements (cm). 51

Table 5.2: Flexible probe lengths at various chimpanzee sites. 55

Table 6.1: Protein, vitamins, and other nutrients in termites, beef, chicken, and fish 61 (/lOOg). lX

ACKNOWLEDGEMENTS

I could not have done this research without the constant guidance ofDrs. Jill Pruetz

(Department of Anthropology, Iowa State University) and William McGrew (Departments of

Anthropology and Zoology, Miami University). Jill, I wish there were someway to repay you for all of your support. Working at Fongoli was the most amazing experience of my life. To

Bill, you are constant inspiration and ultimate guide into the world of primatology, cultural primatology, and specifically termite fishing. Thank you from the bottom of my heart.

Funding was provided by a US Fish and Wildlife Great Ape Conservation Grant awarded to

Dr. Pruetz.

I would like to thank Drs. Matthew G. Hill (Department of Anthropology, Iowa State

University) and Donald Lewis (Department of Entomology, Iowa State University) for methodological advice and thesis layout and editing. For the identification of termites and entomology information, deepest gratitude goes to Dr. Rudolf Scheffrahn, University of

Florida. For assistance in the field I thank Mbouli Camera and Dondo Kante, field assistants for the Fongoli Chimpanzee Research. I appreciate the help of Michel Waller, Fiona Stewart, and Maja Gaspersie. Thanks to Laura Jesse for entomology advice and termite photographic help. For statistical analysis help I thank Norma Leyva (Statistical Consulting at Iowa State

University).

To my family and friends, I could not have done this without your help and support.

Special thanks to: my parents, who have given me a strong work ethic and have always encouraged me to follow my dreams; for your constant support and ear to my troubles,

Jacklyn Ramsey; Kristina Walkup, your time in reading my thesis; Linda Marchant

(Department of Anthropology, Miami University) for your guidance in primatology, you are x like a second mother to me; and finally, to my best friend, Kevin McKee, who has been there through the hard times and has always believed in me. Xl

ABSTRACT

Chimpanzees use tools to fish for termites across Africa, but the ecology of this insectivory has been largely ignored. West African chimpanzees (Pan troglodytes verus) at

Fongoli, southeastern Senegal, fish for termites year-round, with peaks from April to July and in October. Data were collected on 124 termite mounds including information on nearest neighbor mound, shape, habitat-type, percent cover by woody vegetation, height, width, termite activity, chimpanzee activity, and vegetation. The presence of chimpanzee activity was assessed using ethoarchaeological methods, that is, artifacts and remnants l~ft behind were systematically collected and measured. Tool length and plant source were recorded. Of the mounds used by chimpanzees, 59% were in open woodland. Tools were found at 29 mounds (23%), and 20 revisits also proved positive for tools, for a total of 49 assemblages of tools. A total of 401 fresh and recent (younger than 5 days old) flexible probes were analyzed. Results are compared to other sites.

At least six genera of termites occur at Fongoli. Indirect data suggest that the chimpanzees ofFongoli may fish for two of these: Macrotermes and Trinervitermes.

Macrotermes subhyalinus and M bellicosus comprise 75% of the fished mounds. Termite diet may be more diverse for the chimpanzees ofFongoli, given their long fishing season compared to other sites. 1

CHAPTER 1: Introduction

Importance of Chimpanzees to Understanding Human Evolution Studies

One of the principle reasons anthropologists study nonhuman primates is in regards to their role as referential models of human (Homo sapiens) evolution. For example, questions of human evolution initially stimulated research on our closest living relatives, chimpanzees

(Pan troglodytes). Louis Leakey recruited Jane Goodall in the 1960s for this very purpose, to research chimpanzees in Tanzania, at the Gombe field site. Chimpanzees have been studied as referential models for human evolution, not only as analogous models but perhaps more appropriately, as inferential models for the ecology and behavior of hominids.

The chimpanzee ancestor diverged from the human ancestor about 6-8 million years ago, leaving approximately 98% of human and chimpanzee DNA identical (Navarro &

Barton, 2003). Early hominid evolution following this split is thought to have occurred in a savanna-like or open habitat; this contrasts with the standard view of ancestral hominoids living in forests. Thus, studying differences of ecology and behavior in extant primates, such as chimpanzees living in savanna versus forested environments, will provide fresh information for modeling the evolution of the early hominids (Moore, 1996).

Human evolution has been modeled on many different aspects of extant primate behavior, such as diet, habitat, and predation. Research on diet in primates has been especially influential in inferring diet in early hominids. Most dietary studies of chimpanzees have concentrated on the types of fruits and plants these apes consume (see Balcomb et al.,

2000; Matsumoto-Oda & Hayashi, 1997). In regards to diet, one area that has been largely ignored is insectivory. This study concentrates on the ecological context of the insectivorous diet and termite fishing behavior of chimpanzees. 2

Chimpanzees are often referred to as omnivores, much like humans, feeding on everything from seeds to other animals (McGrew, 1983). Although animal prey makes up a low proportion of the chimpanzee diet, fauna provide nutrients, amino acids, and proteins that cannot be acquired from plants (McGrew, 1983; Sussman, 1987). Regarding the faunivorous diet of chimpanzees, mammalian prey especially has been a focal topic of discussion among anthropologists (Stanford & Bunn, 2001). Meat-eating is appealing to anthropologists because it is often associated with hunting and sharing of food (Stanford &

Bunn, 2001). However, invertebrates may be equally important in the chimpanzee diet

(McGrew, 2001). The methods employed by chimpanzees for the consumption of invertebrates are of significance as well, i.e. manipulating plant material into tools to obtain otherwise unattainable insects or insects that sting or bite en masse. Chimpanzees use elementary technology for simple problem solving in order to obtain invertebrate prey

(McGrew, 1992). The ecological details behind this behavior are lacking in termite fishing studies. Here I will examine these details.

Primate Insectivory

Primates are thought to exhibit a "species-specific" dietary pattern (Sussman, 1987).

Most protein obtained by omnivorous primates is acquired through the consumption of insects or leaves (Sussman, 1987). Animal prey does not make up a very high proportion of the primate diet; however, it does seem to be very important (Sussman, 1987). Many studies

often lump vertebrates and invertebrates together regarding their composition in the diet.

However, there may be evidence missed in doing so, such as the importance of each of these

categories, respectively. 3

Insectivory by primates is common. Insects provide high energy and offer many of the same nutrients as vertebrates (McGrew, 2001). According to studies of entomophagy (the consumption of insects) in humans, insects provide more calories (i.e. energy), essential nutrients, and sometimes protein than beef, chicken, or fish (Berenbaum, 1995). Primates consume representatives from most of the orders within the insect class, but there is preference for large-bodied social insects, either wingless or slow-moving (McGrew, 2001).

Obligate insectivores include prosimians, tamarins (Saguinus spp.), and tarsiers (Tarsius spp.) (McGrew, 2001). The latter are the only truly faunivorous primate; however in that they only eat animal prey (Nash, 1984; Warren & Crompton, 1998). Monkeys, such as capuchins

(Cebus sp.), baboons (Papio sp.), langurs (Colobinae), and macaques (Macaca sp.), and lesser apes (Hylobates) occasionally consume insects during opportunistic foraging

(McGrew, 2001). The composition of insects in the diet of these primates can vary however.

Capuchins include up to 20% of invertebrates in their diet (Freeman 1985), while invertebrates account for only 5% of the diet of long-tailed macaques (Sussman, 1987).

Patas monkeys (Erythrocebus patas pyrrhonotus) consume thorn-dwelling social insects from Acacia trees, which make up approximately 36% of their diet, as well as other invertebrates, which comprise another 5% of food items (Isbell, 1998). The special and temporal availability of social insects allows a relatively large-bodied primate to exploit what is usually a costly resource to acquire. Similarly, the insectivory of large-bodied apes is largely limited to social insects.

All of the great apes consume insects and use specialized techniques to obtain them; however, the chimpanzee is the only known non-human primate to habitually use multiple types of tools to acquire insects (McGrew, 1992). Tool use of this nature has been observed 4 across Africa over the many years that chimpanzees have been studied in the wild (Fay &

Carroll, 1994; Goodall, 1963; McBeath & McGrew, 1982; McGrew & Collins, 1985; Suzuki et al., 1995). Chimpanzees are known to be habitual consumers of insects across Africa

(Collins & McGrew, 1985; Fay & Carroll, 1994; Humle & Matsuzawa, 2001; McBeath &

McGrew, 1982; McGrew et al., 1979; Nishida & Uehara, 1980; Sanz et al., 2004; Suzuki et al., 1995). Chimpanzees mainly feed on social insects, such as Hymenoptera (ants, wasps and bees) and Isoptera (termites) (McGrew, 1983) and have been observed to employ various tools in order to obtain them. These patterns of tool use and other behaviors have been termed culture in chimpanzees, a feature once considered uniquely human (McGrew, 2004).

Termite fishing behavior, especially, appears to vary between sites to some degree, but detailed data regarding ecological inferences on such variance are lacking.

Termites as a tax.a are very diverse across Africa (Eggleton et al., 2002). Eggleton and colleagues (2002) found approximately 133 different termite species in only two countries

(Cameroon and Republic of Congo). Termite diversity may depend on the degree of habitat disturbance (Sarr et al., 1998). McGrew (1983) notes that two genera of termites are known to be consumed by chimpanzees: Macrotermes and Pseudacanthotermes. Furthermore, previous studies suggest that the termite genera Odontotermes is not consumed by the chimpanzees but is found at many chimpanzee sites (Collins & McGrew, 1985; 1987). This aspect of chimpanzee diet is little studied.

Non-Human Ape Diet

Diet among the great apes is very diverse. Divergences in diet exist even among sub species (e.g. mountain gorilla as opposed to lowland gorilla). I present here a general breakdown of diet between species of hominoids (Table 1.1 ). Orangutans' (Pon go pygmaeus) 5

diet consists mainly of fruit (54%) as well 54% fruit (Rodman, 1977) (Table 1.1). Western

lowland gorillas (Gorilla gorilla gorilla) at Lope Primate research site in Gabon are

characterized by a diet made up of fruit (46% ), seeds (10% ), leaves (22% ), pith (7%), flowers

(1%), animal prey (5%), and other miscellaneous items (10%) (Tutin et al., 1997). This

differs slightly from another western lowland gorilla site, Bai Hokou in Central African

Republic, where gorillas' diet includes 51% fruit, 26% leaves, 12% bark, and 10% herbs

(Remis, 1997). Table 1.1 averages these two sites for comparison to other great apes.

Mountain Gorillas (Gorilla gorilla beringei) have a specialized diet mainly consisting of

leaves, shoots, and stems (86%) and only 2% is of fruit (Fossey & Harcourt, 1977) (Table

1.1). Chimpanzee (Pan troglodytes) diet is made up of mostly fruit (60%) (averaged from 12

sites including Fongoli) (Pruetz, 2005). Animal prey in the chimpanzee diet accounts for

approximately 8%, but only 2.5% is of vertebrate species (averaged from eight chimpanzees

sites including Fongoli) (Pruetz, 2005). The proportion of the chimpanzee diet consisting of

animal prey is greater than all other non-human hominoids. Insect eating occurs more often

than vertebrate eating; the former is a daily activity while the latter may occur only a few times per year (McGrew, 1983 ). Of the two Orders of social insects commonly consumed by

chimpanzees; Hymenoptera (ants, wasps and bees) and Isoptera (termites) (McGrew, 2001),

the latter will be specifically examined in this study. 6

Table 1.1: Compositions of great ape diets (percentages) Fruit Leaves Flowers Pith, bark, Animal Other and/or THV* Chimpanzees (Pan 62 16 5 9 8 troglodytes) 1 Orangutans (Pongo l 29 2 I 14 pygmaeus)2 ~~J - Gorilla (Gorilla gorilla 49 24 1 10 l~5 11 gorilla)3 --·--~--·-- ~·------L------'------Mountain gorilla (Gorilla 2 86 (includes 2 7 0 3 gorilla beringei) 4 THV*) I I ' I I Averaged from 12 sites, Pruetz, 2005,. 2Rodman, 1977,• 3 Averaged from 2 sites. Tutm et al., 1997; Remis, 1997; 4Fossey & Harcourt, 1977 * Terrestrial Herbaceous Vegetation

Chimpanzee insectivory

Chimpanzees vary across sites in their consumption of animal prey, especially

insects. At Lope, Gabon the chimpanzee diet consists of 9.3% animal prey, including 17

invertebrate and vertebrate species (Tutin et al., 1997). The chimpanzees at Kahuzi-Biega

consume animal prey as 9% of their diet, 4% of which are insect prey consumed seasonally

(Basabose, 2002). Four percent of the diet of Gabon chimpanzees consists of animal foods,

while up to 30% of these chimpanzees' time is spent foraging for insects (Hladik, 1977). The

Gabon chimpanzees consume ant species more than termites, while Gombe, Tanzania

chimpanzees consume more termites than ants (Hladik, 1977). Gombe chimpanzees spend up

to 15% of their active time termite fishing (McGrew, 1983). Elsewhere in Tanzania, Mahale

B group chimpanzees consume Macrotermes with probes and fishing techniques (McGrew &

Collins, 1985), while Mahale K group chimpanzees use both hands and probes for the

acquisition of the termite species Pseudocanthotermes, which can be found in 6.5% of fecal 7 data (Uehara, 1982). Tai and Budongo chimpanzees consume termites, but do not use tools to obtain them (Boesch & Boesch, 1990).

Questions and Goals

The questions addressed explored numerous factors relative to chimpanzee insectivory, particularly regarding termite prey. Based on McBeath and McGrew's (1982) study of Mt. Assirik, Senegal ecology and chimpanzee termite fishing, certain questions emerged. Are Fongoli chimpanzees similar to those studies elsewhere in that Macrotermes is the chief genus consumed by the chimpanzees? Does termite species, in terms of diversity and abundance, influence consumption? If so, do chimpanzees consume the termites that are most abundant, or do other factors influence their choices? McBeath and McGrew ( 1982) addressed questions pertaining to tool material and habitat used to termite fish and found that tools were made from plants growing within two meters of the termite mounds and were often of a single species ( Grewia lasiodiscus ). Would this also be true for the F ongoli chimpanzees? Finally, I expected that the materials used for termite fishing tools at Fongoli would come mainly from the ecotone and surrounding habitat, based on McBeath and

McGrew's findings (1982). "Ecotone" habitats, defined as the transitional zone between

woodland and plateau habitats containing evergreen trees, most often yielded termite fishing

at Assirik (McBeath and McGrew, 1982).

To address some of these questions I replicated many methods used by Mc Beath and

McGrew (1982). These authors, who studied at Mt. Assirik, in Pare National du Niokolo

Koba, Senegal, approximately 45 km from the Fongoli study site, calculated the proportions

of various habitat types in regards to chimpanzee termite fishing. Their goal was to determine

whether habitat influenced the termite fishing behavior of the chimpanzees, and their data 8 supported this conclusion. Habitat use related to chimpanzees termite fishing was also considered in the current study. Termite mounds are at a higher density in shrub savanna habitats when compared to gallery forest (Korb & Linsenmair, 2001 ). During the initial year of study at Fongoli it was found that chimpanzees spent 60% of their observed time in woodland and other open habitats (99% of study area) and 40% in the gallery forest (1 % of study area) (Pruetz, unpublished data). Based on these findings, I asked whether termite fishing occurs most prominently in ecotone habitats at Fongoli. I further examined whether habitat influences termite fishing, and if so, what specific factors within the habitat are the selective pressures. Finally, I compare my findings on the physical properties of the tools used by the Fongoli chimpanzees and the termite species consumed to those findings from studies at other chimpanzee sites.

Study Site and Subjects

The study was conducted at the Fongoli study site in the Tomboronkoto Region, southeastern Senegal, West Africa (12°39 N 12°13 W) (Figure 1.1). Research was conducted from May through August in 2004. This study was conducted in a 40 square kilometer area corresponding to the chimpanzees' core range. Study subjects included the Fongoli community of chimpanzees (Pan troglodytes verus) (Figure 1.2) and termites at the site. The

chimpanzee community at Fongoli is semi-habituated with 18 identified individuals and 27

known community members (Pruetz, 2005). Observations of chimpanzees are based on ad

libitum data following methods from Paterson (2001). Fongoli is important due to its location

in a savanna environment, proposed to be the principal habitat in which early hominids

evolved. The site is considered a mosaic habitat, including woodland, plateau, grassland, and

gallery forest. Early hominids most likely lived in this type of habitat rather than dense 9 forested ones, e.g. Oldowan (Australopithecus and Homo) hominids (Plummer et al., 1999).

Therefore, chimpanzees studied here can provide a referential model regarding the diet and behavior of those early hominids.

Dr. Jill D. Pruetz has been studying chimpanzees at the Fongoli site since April 2001 and has collected extensive data on the composition of the chimpanzee diet here. Her data is prominently from fecal analysis, supplemented by behavioral observations and feeding traces left by chimpanzees. Her data suggest that the Fongoli chimpanzees consume termites most intensively from April to July (Pruetz, unpublished data) but termites are included in the diet every month of the year (Pruetz, 2005). 10

Senegal International boundary Region boundary * National capital Region capital Raliroad Road Track

26 60 75 Kilometers

25 60

8ounda~y reprasantatiOn is _ not necessarily authofitatl_ve. Figure 1.1: Senegal, West Africa (Fongoli Chimpanzee Study Site marked). 11

....·~ ·."Z"_ ...... ~ Figure 1.2: Chimpanzee (Mamadou) of the Fongoli community.

Implications of This Research

Insects have been largely ignored as a major contributor to the diet of chimpanzees.

This study will enlighten primatologists as to the importance of insects in the diet of chimpanzees, which has been argued as more important than mammal feeding (McGrew,

2001). Such significance implies that insects may also have been important to early

(Pliocene) hominids and that they may have used similar technology to obtain such prey. 12

Termites have recently been interpreted as included in the diet of early hominids, specifically Australopithecus robustus, based on evidence from bone tools thought to be used to extract termites from mounds (Backwell & d'Errico, 2001). Chimpanzees use similar wooden tools to perforate termite mounds in order to clear open a ventilation hole for insertion of a flexible probe (Sanz et al., 2004). According to dental traits, gracile

Australopithecines are said to have had a diet much like modern apes today, fairly omnivorous (Teaford & Ungar, 2000). Consequently, it is most likely that hominids and even protohominids used tools before this time (Beck, 1980; McGrew, 1992). However, evidence of these tools in the archaeological record is absent due to decay of organic matter such as wood, twigs, sticks, and leaves. Since tool use is found in most great apes, the assumption may be made that the last common ancestor could also have had some simple form of tool technology (about 6 million years ago). Systematic study of prehistoric diet, concentrating on insects, may allow us to make more direct inferences regarding simple forms of technology like that made from materials used by living non-human apes, e.g. in termite fishing, ant dipping and fishing, and honey extraction.

In support of the hypothesis that insectivory played an important role in early hominid evolution are data from living humans. Many modem day human societies also

consume termites (McGrew, 2001). The Tiv of Nigeria gather caterpillars and termites

opportunistically and will also eat the mineral-rich earth of the termite mounds (Bohannan &

Bohannan, 1968). The Mbuti of Central Africa, consider termites a delicacy (Turnbull,

1965). Insects in the early hominid diet could have provided protein and nutrients that plants

and fruits could not provide, especially if hunting and scavenging were limited. Early

hominids may have been more reliant on insects than large vertebrates. 13

This thesis is laid out in terms of subject; therefore, specific methods for each subject will be presented in each chapter. Chapter 2 reviews literature relevant to this study, while the topics of focus were split into: habitat (chapter 3), termite inhabitants (chapter 4), and termite fishing tools (chapter 5). Conclusions and discussion relating to each topic are included within the chapters and, finally, chapter 6 summarizes each subject and discusses the research as a whole. 14

CHAPTER 2: Significance of Tool Use

"If man is to survive, he will have learned to take a delight in the essential differences between men and between cultures. He will learn that differences in ideas and attitudes are a delight, part of life's exciting variety, not something to fear." - Gene Roddenberry

What is Tool Use?

Tool use was once used as a defining criterion for humanity. More recently, the issue

of culture as a hallmark of our species has been questioned. Primatologists have applied the term 'culture' to non-human primates, namely apes. Culture as a concept in non-human primates was introduced approximately 40 years ago following observations of Japanese

macaques potato washing and rice fishing. Innovations were exhibited by one macaque and these behaviors were adopted by the rest of the troop (Kawai, 1965). This and analogous

behaviors found in chimpanzees (Goodall, 1964) enticed anthropologists to consider these

behaviors within the concept of culture for non-human primates. Recently there has been a

move to compile all "cultural" behaviors for chimpanzees (Figure 2.1) (Whiten et al., 1999;

2002). A typical definition of culture is:

information or behavior - shared by a population or subpopulation - which is acquired from conspecifics through some form of social learning (Whitehead et al., 2004: p.427).

Social learning is included in the definition of culture used by primatologists Whiten and

Ham (1992), as well as teaching and imitation.

One of the most common features of culture in non-human animals is tool use. This

behavior has been suggested for many animal species, but only primates have been observed

to exhibit a wide variety and complexity of tool use. Wild monkeys characteristically do not

show habitual tool-use or manufacture tools as do chimpanzees (van Schaik et al., 1999, but 15

Figure 2.1: Three adult males branch clasp grooming at Fongoli, a behavior that exhibits 'cultural' variation across sites. see Fragaszy et al., 2004). The use of tools is found in all chimpanzee communities thus far studied (McGrew, 1992; Whiten et al., 1999; 2001).

The most widely-used definition of tool use is from Beck (1980: p.10) in which a tool

1s: the external deployment of an unattached environmental object to alter more efficiently the form, position, or condition of another object, another organism, or the user itself when the user holds or carries the tool just prior to use and is responsible for the proper and effective orientation of the tool.

He further stipulated that a tool must be an external object "free of any fixed attachment to the substrate and must be held, carried, or manipulated by the user" (Beck, 1980: p.10,).

There are four modes of tool manufacture: detach (simply separating one object into two and

using the detached object), subtract (removing an object from another to use the latter), add

or combine (two or more objects are connected), and reshape (reconstruction of material to 16 use functionally) (Beck, 1980). Tools can serve many functions: to extend reach, exert more mechanical force, elaborate on gesture in display, control liquids, intimidate predators, function as weapons, and for feeding purposes (Beck, 1980; van Schaik, 1999; Jolly, 1985).

The four levels of tool-use identified according to McGrew and Marchant (1997) are anecdotal, idiosyncratic, habitual, and customary. Anecdotal tool use is a unique or rare event that has the constraints of being an accident, mistake by the subject, or observer error

(McGrew & Marchant, 1997). Idiosyncrasy is an action repeated by one individual and is constrained by limited general ability (McGrew & Marchant, 1997). Habitual tool use is when several individuals repeat an action over time, which may be constrained by context specific explanations (McGrew & Marchant, 1997). Finally, customary tool use is a regular occurrence that all individuals of a community perform although constraints on such use include seasonality, age, or sex-specific groups and may have cross-populational differences

(McGrew & Marchant, 1997). The last two levels are the most intriguing and are considered

'cultural' according to Whiten and colleagues (2001), who have made the first non-human attempt to map out cultural behaviors in non-humans according to similarities and differences across chimpanzee sites. Not all animal tool use qualifies as a cultural behavior according to this recent attempt to interpret behavioral variation (Whiten et al., 2001).

Animal tool use

Birds (Aves)

Numerous bird species have been observed to use tools. The Egyptian vulture

(Neophron percnopterus) has been observed to break open ostrich eggs with stones that are thrown or hammered at the egg (Beck, 1980). The Galapagos finch ( Camarhynchus pallidus), commonly called the woodpecker finch, will hold a twig in its bill to extract insects 17 from holes or hard to reach places; once insects are on the twig, the finch will hold the twig under its foot while it eats the insects from the tool (Beck, 1980). Captive Caledonian crows

(Corvus moneduloides) have demonstrated the ability to spontaneously, without human motivation and mimicry, manipulate vegetation to manufacture a probe for food (Kenwood et al., 2005).

Mammals (Mammalia)

Mammals within a number of different families have been observed to use tools.

Elephants use detach and subtract strategies of manufacture, for example. This includes using twigs or branches held in their trunks to scratch parts of their bodies (Beck, 1980). Asian elephants (Elephas maximus) use branches to repel flies. They modify the branch by holding the stem with the front foot and pulling off side extensions with the trunk (Hart et al., 2001 ).

Polar bears (Ursus maritimus) use a detach method of tool manufacture, but they have also been known to use a subtract method as well. In addition, Polar bears have been seen to use tools for weaponry purposes, where they throw chunks of ice or rocks at walruses or seals, either killing or injuring their prey (Beck, 1980). In Shark Bay, Australia, bottlenose dolphins

(Tursiops truncates) use sponges on their rostra for protection against the rough seafloor.

This is thought to be a foraging specialization, to prevent injury (Rendell & Whitehead,

2001).

Primates and Tool Use

Cebus is the chief tool user among New World primates in captivity (Ottoni &

Mannu, 2001) and in the wild (Boinski et al., 2000). An impressive example involves an

incident where a mother used simple tools (plant materials) to nurse her 6-day-old infant's

head injury (Ritchie & Fragaszy ,1988). Panger (1998) however, indicated that free-ranging 18 capuchins do not exhibit the amount of tool use seen in captivity. In contrast, Phillips (1998) describes a case for tool-use in a wild population of white-fronted capuchins (Cebus capucinus). They habitually use leaves to extract water from tree holes (Phillips, 1998). A recent study found capuchins (Cebus ape/la libidinosus) using stone hammers to crack seeds of Manihot in Caatinga dry forests and stone digging tools for digging up tubers and branches. Twigs were also used for probing of insects, honey, and water (Moura & Lee,

2004). In Piaui, Brazil, Cebus libidinosus has also been observed to use stone hammer and anvil techniques to crack nuts (Fragaszy et al., 2004). Many studies have proposed that capuchin monkeys, alongside chimpanzees, display behavioral variation among populations that cannot be explained by ecological variables (Fragaszy et al., 2004, McGrew &

Marchant, 1997; Panger, 1998; Visalberghi & McGrew, 1997).

Few species have been known to use feeding tools in the wild, but among those that do are: Cebus spp., Macaca spp., Papio spp., Pongo spp., and Pan spp (van Schaik, 1999).

The capacity for tool-use is thought to have evolved three times in evolutionary past, in capuchins (Cebus), cercopithecines, and apes, which are taxonomically separated and not closely related (van Schaik et al., 1999). Van Schaik et al.'s (1999) predictive model for determining likely tool-users in feeding behavior positively correlates such behavior with extractive foraging tasks, intelligence, dexterity, gregariousness, and social tolerance within primates. A primate is more likely to use tools when all of these factors are present.

Ape tool use (Hominoidea)

Great apes (Superfamily Hominoidea) are the only known non-human animals to use all four forms of tool manufacture as defined by Beck (1980). It is rare to find tool-use among the lesser apes, gibbons and siamangs (Hylobates spp.) (Beck, 1980). The unique 19 reliance on tools for specific tasks by great apes is one trait that initiates the great interest in ape behavior by anthropologists. Only studies of wild apes will be reviewed here. There is no doubt that humans are complex and intelligent apes, but to understand how we arrived here can only be determined by understanding human evolution. The apes are our closest living relatives and can help guide us to possible answers.

Orangutans (Pongo pygmaeus)

Orangutans are the only habitual tool-users among primates besides chimpanzees.

Orangutans use tools arboreally with their mouths (Fox & Muhammad, 2002). Suaq

Balimbing orangutans use sticks or twigs to extract honey from stingless beehives in trees

(van Schaik, 1999). The Suaq Balimbing orangutans also consume more insects than orangutans at other sites (Fox et al., 2004). Using short twigs, these orangutans also take seeds from Neesia fruits, which are protected by stinging hairs. The twigs are modified by breaking and stripping off leaves or chewing on the twig end, fraying it. Twigs are used as both a wedge and a spoon (van Schaik, 1999). Orangutans also extract termites and ants from tree holes (van Schaik et al., 2003a) using unmodified twigs.

An orangutan cultural database was recently initiated, modeled after that of Whiten and colleagues for chimpanzees (1999; 2001). Van Schaik and colleagues (2003b) compiled a list of behaviors interpreted as cultural using a similar definition to the one described earlier for chimpanzees. Some of the tool use behaviors include leaf wipe (wiping face with leaves), sun cover (building a cover on nests), scratch stick (using detached stick to scratch body parts), branch as swatter (leafy branch swatting at bees/wasps), tree hole tool use, seed extraction tool, and branch scoop (drinking water extracted from a tree hole with a leafy

branch) (van Schaik et al., 2003b). 20

Gorillas (Gorilla gorilla)

A few instances have been described for tool-use among wild gorillas; however, none are interpreted as habitual for any community. Mountain gorillas have been said to use just as much skill with their hands, if not more, in processing thistles than is seen with chimpanzee tool use. Gorillas, according to Byrne and colleagues (2001 ), do not have any need for tools.

Byrne and colleagues (2001) suggested that this complexity in processing food shows that gorillas have a great cognitive capacity.

Bonobos (Pan paniscus)

Little is known about the tool-use behavior of bonobos (Pan paniscus) due to the lack of study of wild populations. In the wild, leafy twigs are used to cover the body from the cold

(Lomako) and twigs are used to shelter nests/beds (Wamba) (Fruth et al., 1999). Branches are also used in display behaviors. In the wild, bonobos display little material culture when compared to chimpanzees, however (Fruth et al., 1999).

Chimpanzees (Pan troglodytes)

The chimpanzee (Pan troglodytes) is the only extant non-human primate known to use a tool-kit (McGrew et al., 2003). A tool kit is the sum total of tools used or the repertoire possessed by a population, while a tool set is the sequential use of tools. An example of both is when chimpanzees probe for honey: they first use a stick chisel; a slimmer chisel is used next; third, they puncture a hole, and finally use a dip stick to retrieve the goods.

Chimpanzees exhibit a wide variety of tool using behavior (Beck, 1980; Whiten et al.,

2001). They have been observed to do everything from throwing (aimed and unaimed) to nut

cracking, using the same hammer and anvil technique that hominids have been

characteristically said to exhibit. Chimpanzee tool-use can be divided into three categories: 21 social (antagonistic and non-antagonistic), subsistence, and maintenance (McGrew et al.,

2003). To illustrate the diversity among chimpanzees, Tai, Ivory Coast chimpanzees have 20 different tools in their tool kit, while Budongo, Uganda chimpanzees have only five. There are no ecological explanations as to why tools of the type used by Tai chimpanzees do not appear at Budongo (Whiten et al., 2001).

Under certain environmental conditions (i.e. dry forests at Bossou, Guinea), wild chimpanzees depend on tool use for the acquisition of reliable food resources (Y amakoshi,

1998). Subsistence technology is most representative of chimpanzee 'culture'. Of 39 behavior patterns judged to be cultural, 22 are used for subsistence (Whiten et al., 2001 ).

Chimpanzees spend 15% of their daily time budget at some sites using tools to fish for termites and an equal amount oftime nut cracking at other sites (McGrew, 1992).

Termite Fishing

My research concentrates on termite fishing for several reasons. First, termites are a substantive food that may exceed in quantity all other vertebrates consumed in the rainy season (McGrew, 2003). Second, termites provide a substantial source of nutrition. The use of tools for termite fishing will be the main target of this study.

Gorillas and bonobos do not use tools for termite fishing (McGrew, 2001).

Orangutans use probes for bees and honey, but use a ' smash and grab' technique for termites, ants, wasps, and bees (van Schaik, 1999; van Schaik et al., 2003a). Chimpanzees customarily use probes to fish for termites (Figure 2.2) or ants. Both chimpanzees and orangutans exhibit a female dietary preference for insects and a male preference for mammalian protein intake

(McGrew, 2001). When chimpanzees customarily use probes to fish for termites or ants, these activities give them many more kilocalories than the energy invested in the activity 22

(McGrew, 2001). Thus, termites are a high-value food that is obtainable using a simple technology.

Termite fishing involves using probes made of twigs, stems, bark, leaves, or vines; these are then inserted into a ventilation hole of a termite mound; termites attack the intruding object, and the chimpanzee carefully withdraws the probe from the hole and eats the affixed termites (Goodall, 1963; 1968). Termite fishing is habitual at the Assirik, Senegal site and customary at Gombe and Mahale (Whiten et al., 2001). This pattern is also found at the Fongoli, Senegal site and is the focus of the current study.

Termite fishing is practiced at many sites across Africa: Assirik, Senegal (McGrew et al., 1979; McBeath & McGrew, 1982); Rio Muni, Equatorial Guinea (McGrew et al., 1979);

Gombe, Tanzania (McGrew et al., 1979); Mahale, Tanzania (McGrew & Collins, 1985);

Ndakan, Congo (Fay & Carroll, 1994); Ndoki, Congo (Suzuki et al., 1995); Goualougo,

Republic of Congo (Sanz et al., 2004). McBeath & McGrew (1982) found that, at Mt. Assirik the majority of termite fishing occurs in an ecotone habitat. This is the zone between plateau and woodland where water drainage produces an area characterized by evergreen vegetation.

They concluded that habitat was influential on this behavior. Such ecological influences are important to consider when attempting to assess the cultural versus ecological aspects of non human primate behavior. 23

Figure 2.2: Chimpanzee termite fishing.

Implications for Culture?

Tools are considered an important part of the chimpanzee cultural database proposed by Whiten et al. (1999, 2001). Such cultural implications are remarkable, considering humans were thought to be the only extant animal to possess culture for many decades. The only known animals to habitually use different and multiple tools are chimpanzees and humans (Boesch, 2003). Therefore, research on chimpanzee tool use may provide referential models for hominid (early human) tool evolution where fossil evidence is lacking:

Five types of evidence are used in order to make inferences about hominid life: fossilized bones and teeth, artifacts, ecofacts, etc.; flora and fauna fossilized in association with hominid dwellings; extant traditional societies (i.e. hunter-gatherers); and behavior of extant non-human primates. Tools first appear in the hominid fossil record about 2.5 million years ago, although new reports may suggest even earlier evidence (see Blackwell & d'Errico, 2001 ). It is still debated as to whether the first tool users were Homo or

Australopithecus. However, it is likely that hominids and even protohominids used tools 24 before this time (Beck, 1980; McGrew, 2001). The evidence for particular tools in the fossil record are absent due to ecological decay of organic matters such as wood, twigs, sticks, and leaves.

Regarding the evolution of tool-use, it appears that, with an increase in intelligence and plasticity, there was an associated increase in the technology of tools (Rightmire, 2004).

Humans have more interest in objects for the purposes of tool use in all areas, whereas non human primates use tools primarily for feeding purposes. No one can deny that humans have a superior knowledge and use of tools and technology than any other species on earth. Yet we cannot ignore where that knowledge has come from and that we are not as unique as we may think. Human technology has evolved to the level of computers and airplanes. Our technology has had a massive impact on the biological and physical environment to an extent of which we are only starting to become aware. It may be ironic that tool-use is considered one of the main shapers of our species but may ultimately may end up being our downfall

(Beck, 1980). Perhaps to understand this eventual destruction we have to examine how tool use was created.

Whitehead and colleagues (2004) suggest that cultural diversity should be a promotion for conservation efforts in all animals (see also Norris, 2002). For example, the evolution of culture is proposed to be very important in the Order Cetaceans due to the constant change in marine ecosystems, allowing whales and dolphins to be able to move monthly (Rendell & Whitehead, 2001; Norris, 2002). Cultural behavior in terms of travel routes provide a way to cope with the changing environment as well as a patchy food resource distribution (Rendell & Whitehead, 2001; Norris, 2002). Cetaceans also follow 25 general rules for models of culture: large brain, long gestation, sexual maturity age, and long lifespan (Rendell & Whitehead, 2001 ).

Culture as we know it may exist all around us, not just in our own species, but in others as well. Why has human culture evolved to what it is? Does it flaunt our superior intelligence? Or does it disprove our intelligence if we ultimately destroy ourselves? If chimpanzees have culture it cannot be known for certain. This lies in the minds of our subjects and it is left to primatologists to make inferences. However, with the increasing information from not only chimpanzees but also other primates, whales, and dolphins, this possibility is difficult to ignore. There is much to justify the role of culture studies within primatology, resulting in a new subfield: cultural primatology.

Tool use is an important component in cultural studies. This research attempts to examine specific features of tool use in chimpanzees. The habitat and vegetation, termite diversity and abundance, and characteristics of tools are examined here to further our knowledge of the ecology of termite fishing. 26

CHAPTER 3: Ecology of Study Site

Introduction

Senegal, West Africa is located in the region stretching across northern Africa from the west (Senegal) to the east (Ethiopia) known as sub-Saharan Africa. Southeastern Senegal is described as having a long dry season (November to April/May) and a short wet/rainy season (May to October) (Carter et al., 2003). Temperatures at Fongoli reach their highest peaks in the dry season (>37°C) (Pruetz, unpublished data). This region represents a semi arid, open environment between the Sudanian and Sudo-Guinean vegetative belts (Carter et al., 2003). It has been suggested that chimpanzees in dry habitats are more affected by seasonality than chimpanzees living in more forested habitats (Hunt & McGrew, 2002). This led to the question of how such a habitat would affect the chimpanzee insectivorous diet in terms of context. Would habitat also affect where chimpanzees eat?

The Fongoli site is similar to that of Assirik, with a habitat that is more open compared to other chimpanzee study sites in Africa (McGrew et al., 1981 ). Baldwin and colleagues (1982) found the Mt. Assirik chimpanzees to utilize woodland and forest habitats

significantly more than plateau habitats, which were mainly used for travel. Seasons affected

use of habitat, when temperatures rose in the dry season, vegetation coverage and water

resources became more important for the chimpanzees at this time (Baldwin et al., 1982;

Tutin et al., 1983). This need for water and shaded areas was interpreted as correlated with

thermoregulation issues in the dry season, but was not thought to be related to food

availability (Baldwin et al., 1982; Tutin et al., 1983). Fongoli's chimpanzees, 45 km

southeast of Mt. Assirik, are described as using habitat comparable to that of Mt. Assirik

(Pruetz, unpublished data). Preliminary data indicate similar selective use of certain habitats. 27

During the initial year of study, at Fongoli it was found that chimpanzees spent 60% of their

time in woodland and other open habitats (99% of study area) and 40% in the gallery forest

(1 % of study area) (Pruetz, unpublished data). Given the ecological similarities at these two

sites, chimpanzee behavioral similarities are also expected.

McBeath and McGrew (1982) found that habitat at Mt. Assirik influenced

chimpanzees' termite fishing behavior. It was hypothesized that habitat would be influential to the eating habits of the chimpanzees, and that habitat would most likely provide the vegetation needed for making tools and for coverage from the sun. Ecotone habitat, defined as the transitional zone between woodland and plateau habitats, was the habitat that yielded most termite fishing (McBeath and McGrew, 1982).

Methods

To address the possibility that habitat use may influence termite fishing at the Fongoli

site I replicated many methods ofMcBeath and McGrew (1982). Based on their findings, the

question of whether termite fishing occurs most prominently in ecotone habitats is addressed

here. Additionally I asked if habitat influences termite fishing, and subsequently, what

specific factors within the habitat are the selective pressures promoting this behavior.

F ongoli is a mosaic habitat much like that of Mt. Assirik (McGrew et al., 1981; Tutin

et al., 1983) but different from most other chimpanzee sites (Pruetz, 2005). The site includes

woodland (Figure 3.ld), bamboo woodland, plateau (Figure 3.lb), grassland, ecotone (Figure

3. lc), and gallery forest (Figure 3. la). Definitions of the habitat follow Pruetz et al. (2002;

Table 3.1). Open habitat includes grassland, plateau, and woodland, following Pruetz (2005)

where open habitat is defined by the nature of the canopy and vegetation coverage. Closed

habitat includes ecotone and gallery forests habitats. 28

Table 3.1: Definitions of habitat types (from Pruetz et al., 2002) Bamboo · "Flat-leaved savanna with isolated palms and deciduous trees". Ecotone "Where water runoff from plateau edge produces a strip of evergreen woody vegetation". Gallery forest ---+-'-'T_r_o~pic_a_l _se_m__ i-_ d_e_c1_·d _u_o_u_s_lo_w_ la_n_d_fi_o_r_es_t_" .______Grassland "Narrow-leveled savanna with isolated deciduous trees". Plateau ''Narrow-leveled savanna with isolated deciduous shrubs". --~------Woodland "Drought-deciduous lowland woodland".

Figure 3.1: Habitats ofFongoli a.) Gallery forest, b.) Plateau, c.) Ecotone, d.) Woodland.

Data collection

Eight transects, each approximately one kilometer in length, were surveyed to assess habitat and termite mound distribution. Any termite mound found within 20m (total transect 29 width of 40m) either side of the transect line was recorded along with habitat to determine mound density. The habitats were recorded for analyses of termite mound density according to habitat. Termite mounds (both unused and used by chimpanzees) were marked with the assistance of field assistants, and mapped using a Global Positioning System (GPS: Magellan model). A total of 124 mounds were marked and monitored. To determine chimpanzee activity at termite mounds, indirect evidence was used according to ethoarchaeology methods following McGrew et al. (2003). Characteristics recorded included vegetation around mound, percent vegetation coverage of mound, habitat, characteristics of mounds (shape, width, height, and activity) evidence of tools use (discarded probes used by chimpanzees), and distance to nearest neighbor mound.

Analyses

Every change of habitat along each transect was used in the estimation of habitat distribution. Mound density was estimated using transect data and habitat proportions; a grid of 1OOm by 1OOm (1 hectare) was used to estimate the relative densities of termite mounds.

Nearest neighbor methods (following Cottam & Curtis, 1956) were used to give an estimate of total mound density at the site; this method could not be used within each habitat due to low sample size. Finally, analyses were done of characteristics on mounds used by chimpanzees for termite fishing to compare with all mounds sampled.

Results

Mound distribution is shown in Figure 3.2. Termite mounds were marked and monitored within the core study area of 40 square kilometers at Fongoli. The average

distance between mounds is 29 meters, with a range of 2-85m. ·!,C l~ 1 · \ L ." •

: :~ ~

•2' 41 : : ------. ------._ l: ...... _ ...... ~: : :_, "'""' ' . # - ~H "~-mn"m+ ' ... ' ' *~ " ' : #'m' i : : • ~ - ! ~ : i& . ' •~ ,~-w l •2'~j ' .. "·------r------~------. ------+------!--

..:~ •'tlZlllm

1; •3; . : ~ ------:------: -- U l nmO " nn~~~ nmmnrnn HUTU n '"' ' ' n• nm :. i : • -- - - - ' • • • nn•

' ' 2km

·--=-----~~-"-=··-==·-~----~----~-~~-==""'..--·==--~ -·------=~-=----=~--=--==--·--===·-----~---~---==~- ·--=----~=~=------=~~--'-=·--=:~-=--==~=------·--·--- -='~~=-~---=·..···-=--~----~--- Figure 3.2: Recorded termite mound distribution on GPS Magellan (UTM units).

w 0 31

Transect data revealed a habitat that is diverse. Open woodland accounts for the majority of area (32%) (Figure 3.3). Grassland and plateau each accounted for 24% of the total area sampled (Figure 3.3). Another 13% was bamboo woodland (Figure 3.3). Both ecotone and cultivated fields accounted for small percentages of the site (Figure 3.3). Finally, gallery forest is only 1% of Fongoli's surface area (Figure 3.3). These findings are comparable to Pruetz' (2005) findings using quadrat methods.

Of the 124 mounds marked and monitored, 43% were in open woodland. A total of

49 tool assemblages at 29 mounds were recorded. Mounds used by chimpanzees were found in woodland (59%), ecotone (21 %), bamboo (10%), and gallery forest (10%); none were found in grassland or plateau (Figure 3.4). Chimpanzees termite fish in woodland significantly more than in other habitats (X2=18, df=3, p-value=0.0004, two-tailed). Gallery and ecotone habitats are also used by Fongoli chimpanzees for termite fishing more than expected given the availability of these habitats (ecotone X2=7.68, df=6, p-value=.0014, two- tailed; gallery X2=23.04, df=6, p-value=<.0001 , two tailed).

Reid 3°/o Grassland W:xrland 24% 32%

Ecctone v 3%

Figure 3.3: Fongoli Habitat distribution. 32

GO% -+------__, D Fongoli habitat availability • LSed by Oiirrpan2E!e5 50%

§ 40% :e 8. 30%

~ 20%

10% //0% '

Figure 3.4: Habitat distribution at Fongoli and location of termite mounds used by chimpanzees.

Termite mound density (based on transect data) was the highest in plateau and grassland habitat, numbering approximately 10 mounds per hectare (Table 3.2). Ecotone was characterized by the next highest termite mound density, followed by woodland and bamboo.

Cultivated fields yielded fewer mounds per hectare at Fongoli, but gallery forest exhibited the lowest termite mound density, with approximately one mound per hectare. Figure 3.5 demonstrates the habitats used by chimpanzees to termite fish in relation to the mound density proportions at Fongoli. 33

Table 3.2: Mound density (per hectare at Fon oli for each category of habitat. Mound density ---~,....__..,~er hectare) Woodland 7 Ecotone 8 -----4------··-----··-- Gallery 1 Bamboo 7 Plateau 10 -----il------Grassland 1O Field 5

700/o 00% -t---==------j. f\/bn:j density i_ o Used by Olirrpanzees I 50% 6 4(1>/o 300/o ia. 20% 100/o 00/o 0-0 / # Habitat Figure 3.5: Termite mound density proportion versus termite fished habitat used by chimpanzees at Fongoli.

Nearest neighbor analyses indicate that the mean distance between two termite mounds in all habitats is 28.8 meters, with an expected value of 9.5 m based on nearest neighbor analysis (Cottam and Curtis, 1956). According to a Poisson analysis, this finding is

significantly higher than expected (Poisson, p-value=0.0018, two-tailed). Therefore, termite

mounds are more widely distributed than expected for Fongoli. Lee and Wood (1971) 34 however, state that nearest neighbor analyses are not appropriate measures of termite mounds because they are not randomly distributed.

Discussion and Conclusions

The Fongoli study site consists mainly of open habitat (including woodland), which limits the chimpanzees' availability of water and thermoregulation abilities especially in the dry season. Ecotone habitat, a closed habitat, was characterized by 21 % of the total chimpanzee fishing assemblages. Ecotone only makes up 3% of the site. Gallery forest habitat was used in 10% of termite fishing instances and only makes up 1% of area (Figure

3 .1 ). Both of these habitats were used significantly more than what would be expected at

Fongoli. In contrast to Mt. Assirik (McBeath & McGrew, 1982), the Fongoli chimpanzees did not use ecotone most often for termite fishing; instead they used open woodland. This may be due to the reduced proportion of ecotone at Fongoli.

The termite data presented here indicate that Fongoli chimpanzees are using the habitats with low to medium densities of mounds for termite fishing. Therefore, termite mound density does not influence the termite fishing behavior of the Fongoli chimpanzees.

The termite mound density at Fongoli is characteristic of an open mosaic habitat (Korb &

Linsenmair, 2001; Lee & Wood, 1971). Entomology studies indicate that Macrotermes bellicosus has a higher density in open habitats in Cote d'Ivoire and lower densities in the gallery forests (Korb & Linsenmair, 2001). It has been estimated that Macrotermitinae have a density ofless than 10 mounds per hectare in Africa (Lee & Wood, 1971). Collins and

McGrew ( 1987) estimated the density of termite mounds using transect data at Mahale. The

density of termite mounds was 10.6 mounds per hectare at Kasoje and 13.5 mounds per

hectare at Bilenge, with a total of 15.6 mounds per ha overall in open woodland. Fongoli 35 termite mound density is comparable to that found at Mahale, although the habitats are different. Mahale is characterized as having denser forest (Collins & McGrew, 1987), compared to the mosaic of open habitats at Fongoli. At another Tanzanian site of Gombe, mound density was only 5.6 per ha. Collins and McGrew (1987) attribute this low density to the steep slopes found in Gombe. No other chimpanzee site provides data for comparison.

Senegal's open environment has been used in reference to understanding early hominids behavior and ecology during the Plio-Pleistocene (McGrew et al., 1981 ). Habitats in eastern Africa are more forested and comparable for adaptations of Miocene primates in a forest-woodland habitat (Collins & McGrew, 1988). Data presented here provide evidence for the selective use of woodland and forested habitats by the chimpanzees for a food resource (termites), in comparison with grassland and plateau habitats. Thus, habitat rather than termite mound density influences chimpanzee termite fishing. Mound density is high in plateau and grassland habitats, where the chimpanzees do not consume termites (Figure 3.5).

However, termite species diversity does differ between habitats, and this could also be a factor in chimpanzee termite feeding behavior. I address this issue in the following chapter. 36

CHAPTER 4: Fongoli Termites

Introduction

Here I present an in-depth analysis of the termite species, their mound characteristics, and their importance in the diet of the Fongoli chimpanzees. McGrew (1983) noted that mainly two genera oflsoptera (termites) are consumed by chimpanzees: Macrotermes and

Pseudacanthotermes. Previous studies suggest that the termite genus Odontotermes is not consumed by chimpanzees, although they are found at several chimpanzee sites (Collins &

McGrew, 1985; 1987). Most sites only report the species that are included in the diet of the chimpanzees (Table 4.1 ). This study went further to record all termite species at the site in order to determine availability of the species chimpanzees select for consumption. The only comparable data come from Collins and McGrew (1985; 1987) for the Tanzanian sites of

Mahale Bilenge, Mahale Kasoje, and Gombe.

The questions addressed here explored numerous factors relative to insectivory of chimpanzees, particularly termites. Based on McBeath and McGrew (1982) questions included: Is Macrotermes the chief genus consumed by the Fongoli chimpanzees? Does termite species, in terms of diversity and abundance, influence consumption? If so, do chimpanzees consume the termites that are most abundant, or do other factors influence their choices?

Methods

Multiple termite samples were collected at mounds with tools and at non-tool mounds. A total of 39 termite samples were collected using chimpanzee tool extraction techniques and forceps (Figure 4.1 ), then preserved in vials of 85% ethanol. An attempt was

made to sample all species and castes (workers, immatures, reproductives, and both minor 37 and major soldiers) of termites. Dr. Rudolf Scheffrahn from the University of Florida identified the samples.

Termite species are reviewed in terms of habitat, abundance, and whether chimpanzees consume them. Mound characteristics are analyzed according to mound shape and height. Shape categories were innovated by the author to include convex, castle, mushroom, and miscellaneous. Convex is a mound that is dome shaped and may have satellite mounds, smaller mounds less than Sm distant (Figure 4.2a). Castle mounds can be several meters tall, with intricate towers, and resembling a castle. Mushroom mounds are low mounds less than one meter in height with a narrow stem that domes out at the top (Figure

4.2b, c ). Miscellaneous mounds are those mounds that did not seem to take a particular shape or were on a slope, which resulted in a misshapen structure. Some structures can have multiple shapes; these generally had double inhabitant genera (Figure 4.3). In these cases the dominant structure was used for analyses.

Figure 4.1: Author and Mbouli Camera (field assistant) collecting termites. 38

Figure 4.2: Mounds at Fongoli a) convex mound under Saba vine in ecotone habitat b) mushroom mounds on plateau c) Mushroom mound close up

Figure 4.3: Double inhabitant mound (convex with mushroom) in woodland habitat. 39

Results

Mound characteristics

No evidence was found to suggest that Fongoli chimpanzees ever fished at the mushroom shaped mounds; instead all mounds used had convex, castle, or miscellaneous shapes. Convex mounds were used significantly more by the chimpanzees than all other shapes (X2=20, df=2, p=0.0001, two-tailed). All termite mounds recorded for the study had an average height of 97 cm. Fished mounds used by chimpanzees had an average height of

100 cm, and a range of 25-215 cm.

Termite distribution

Termite species varied within the study site: at least six different genera of termites were recorded at Fongoli: Cubitermes (2 species), Microcerotermes (1 species), Amitermes

(1 species), Trinervitermes (3 species), Macrotermes (2 species), and unidentified species of

Termitinae (Table 4.1 ). Cubitermes are mostly found associated with mushroom shaped mounds on the plateau and grassland, while Macrotermes are associated with a convex mound shape and are generally in woodland or forested habitats. However, Trinervitermes is likely to be found in all habitats and associated with convex or mushroom shaped mounds.

Termites were collected at 29% of all mounds recorded; another 7% were abandoned of termite inhabitants and 64% were not identified. Most of the termites found at Fongoli were Macrotermes sp (36%) and Trinervitermes sp (36%), followed by Cubitermes,

Microcerotermes, and Amitermes (Figure 4.4). In 18% of cases, more than one termite genus inhabited a mound. Two genera and four species were found in association with termite tools: Macrotermes subhyalinus, M bellicosus, Trinervitermes togoensis, and T geminatus.

These termites were sampled at mounds where tools were found. Multiple species inhabiting 40 termite mounds and few direct observations of termite fishing made it difficult to determine which termite species are being consumed by the chimpanzees. However, Macrotermes species were most often (75% of cases) found in association with tools. There were three instances where Trinervitermes was associated with tools, but two of these were double inhabitant mounds with Macrotermes also resident. Figure 4.5 illustrates termite soldiers of the species associated with tools.

Amitermes Microcerotermes 2% 7%

Cubitermes 19%

Trinervitermes 36%

Figure 4.4: Termites diversity at Fongoli. 41

Figure 4.5: Termite soldiers of the genera associated with tools a) Macrotermes bellicosus b) Macrotermes subhyalinus c) Trinervitermes occidentalis d) Trinervitermes geminatus

Discussion and Conclusion

The Fongoli chimpanzees may rely on termites more than chimpanzees at other sites.

They are one of a few known wild populations to eat these social insects year-round. Data from fecal analyses indicate termites are present in the chimpanzee diet in all months of the year (Pruetz, 2005).

Both Macrotermes spp. and Trinervitermes spp. were found in association with tools at Fongoli. However, since there are instances of multiple inhabitants at a mound, it cannot be conclusively said that Trinervitermes are eaten by the Fongoli chimpanzees. Further 42 micro-fecal analyses could remedy this lack of evidence. Other sites have reported the consumption of multiple termite genera however. Table 4.1 lists the termite species known to be consumed or found in association with termite fishing tools at sites across Africa. The

Mahale B group seems to select Macrotermes from a number of possible species; however the Mahale K group consumes Pseudacanthotermes springier apparently due to the absence of Macrotermes in the area (Collins & McGrew, 1985). At Gombe, the chimpanzees fish for

Macrotermes and use their hands to opportunistically acquire Pseudacanthotermes (McGrew,

1992). Fongoli chimpanzees could use this same technique with Trinervitermes when at a mound that houses both this species and Macrotermes. More data are needed to determine conclusively which termite species are eaten by Fongoli chimpanzees. In fact, Fongoli chimpanzees have been observed to use tools to termite fish as well as their hands to capture termite prey at the same mound (Pruetz, personal communication).

Termite fishing in the dry season at Fongoli may be related to the behavioral characteristics of Trinervitermes. Trinervitermes are more active in the dry season (Ohiagu,

1979), when Macrotermes is more difficult to obtain from mounds. Trinervitermes are a

grass-harvesting termite (Ohiagu, 1979), which may provide chimpanzees with certain

nutrients that wood-eating Macrotermes do not provide (Pearce, 1997). Effects on the soil

acidity and moisture or the activity of the termites may be affected by having double

occupants such as Trinervitermes and Macrotermes. Different kinds of soil are used by

different termite species. Also, Trinervitermes often forages on the surface (Ohiagu, 1979),

which provides easy access for chimpanzees. Trinervitermes secretes a toxic substance that

immobilizes other insect enemies (Lee & Wood, 1971); whether or not this toxin is secreted

on tools and whether it affects an animal the size of a chimpanzee is unknown. Table 4.1: Termite genera and species at chimpanzee sites across Africa. Genus Species Fongoli, Mt. Assirik, Gombe, I Mahale - j Mahale- Budongo Rio Muni2 Senegal1 Senegal2·4 Tanzania2·3· Bilenge, Kasoje, Forest, Bossou8•14 13 I Tanzania3,s Tanzania Uganda7 Ndakan9 3,6,12 ' Ndoki10 I Goualougo, I I I Congo11 p c p c I p c p c I p c p c p c I ' I I I I Cubitermes sankurensis x x x l I i I I loubetsiensis x I I I I I I I I I I I Otherspp? I x i I x I X(-) I Microcerotermes I parvus or other x x I x I I I i SOD. I I I I I I I I I I I I ! Macrotermes bellicosus x I x I I x I x I I I subhyalinus x x x x I i Otherspp? I I I I I I x I x I x I I x I x Trinervitermes geminatus I x x I i I I i l occidentalis I x I I I I I I I togoensis x x I I I I Amitermes evuncifer x I I I family Termitidae spp? x i x I x x I ! I I I I I I Pseudacanthotermes spiniger or spp? x I X(-) I x x X(-) I I I I i I Odontotermes spp? I x x I x I I I I I i I I P = Present; C = Consumed with or without tools (-) or found in association with tools 'Results of this study; 2McGrew et al., 1979; 3Collins & McGrew, 1987; 4McBeath & McGrew, 1982; 5Collins & McGrew, 1985; 6McGrew, 1983; 7Newton-Fisher, 1999, 8Humle & Matsuzawa, 2001 ; 9Fay & Carroll, 1994; 10Suzuki et al., 1995; 11 Sanz et al., 2004; 12Nishida & Uehara, 1980; 13McGrew, 1992, 14Sugiyama & Koman, 1987.

w.i:::.. 44

Collins and McGrew (1987) found 16% of termite mounds in Gombe, Mahale

(Bilenge), and Mahale (Kasoje) to have more than one species of termites in residence.

Another 6.5% of mounds had more than two species resident. Table 4.1 describes termites known to be present and eaten at chimpanzee sites. Only Collins and McGrew (1987) presented comparable data on termite species abundance. They compared Kasoje, Bilenge, and Gombe, Tanzanian sites. Kasoje and Gombe were characterized by a greater percentage of Pseudacanthotermes, while the most common termite species at Bilenge was

Odontotermes (Collins and McGrew, 1987). At all sites, Macrotermes were uncommon, which may explain the defined seasonality of termite fishing at these sites. At Fongoli, however, Macrotermes is one of the most common termite genera at the site, which may explain the chimpanzee's year-round termite fishing. Collins and McGrew (1987) found that termite species' distribution affected chimpanzees' termite fishing, which may also be true for Fongoli. The Fongoli chimpanzees termite fish in woodland more than any other habitat, which may potentially be correlated with Macrotermes distribution in woodland habitats.

Thus, habitat does influence termite fishing at Fongoli. Although, seasonality was found to be important at Mt. Assirik, termite abundance and distribution were not calculated

(McBeath & McGrew, 1982).

Termite consumption by Fongoli chimpanzees is an important factor when compared

to other sites, due to Fongoli's year-around termite fishing behavior. In terms of food scarcity

at Fongoli, the period oflowest fruit availability is July- August (Pruetz, 2005). Fruit scarcity

does not seem to correlate with termite abundance in the diet (Pruetz, unpublished data).

Termites therefore, may not be a fall-back food resource for Fongoli chimpanzees. Rather, 45 the chimpanzees consume termites throughout the year at this site, with a few peaks from

April to July and in October (Pruetz, unpublished data).

At other sites such as Mt. Assirik (McGrew et al., 1979), Gembe (McGrew et al. ,

1979) and Mahale (McGrew & Collins, 1985), seasonality in termite fishing coincides with the initial months of the wet season. At Okorobiko, Rio Muni, there is no seasonality in this behavior, and termite fishing tools are found throughout the year (McGrew et al., 1979).

Ndoki forest, Congo, also exhibits a lack of seasonal variation in termite fishing behavior, and there was no correlation with rainfall in the months that provided the most tools (Suzuki et al., 1995). Suzuki and colleagues (1995) concluded that the termite soldiers at Ndoki must be available to eat throughout the year for the chimpanzees. The same seems to be true at

Fongoli.

Eggleton et al. (2002) recorded at least 133 different termite species in western

Africa. IfFongoli chimpanzees are only consuming Macrotermes, they may be exploiting only a small proportion of the known termite species found here. More research on relative abundance of termite species needs to be conducted. Further investigation into whether or not the chimpanzees are consuming only one of the species inhabiting a mound also needs to be done. If two species are present, yet only one is consumed, this would then suggest that the chimpanzees are selective regarding the termites they consume. Also, as a caveat, the observed indisputable evidence of termites in the diet of chimpanzees is the hard mandibles and heads of the Macrotermes soldiers in feces. The consumption of workers and possibly other species of termites is difficult to discern based on this type of macro-analysis. More

extensive methods that sample termite diversity at a site will give a better understanding of

the relation between termite availability and consumption, especially related to seasonality. 46

Micro-analysis of termite remains in the feces may help to give more detailed results of chimpanzee invertebrate diet. 47

CHAPTER 5: Chimpanzee Termite Fishing Tools

Introduction

Termite fishing is labeled a 'cultural' behavior by primatologists and has been observed at many chimpanzee sites across Africa (Whiten et al., 2001 ). This behavior set a major course in anthropology, where tool use was once thought to be a uniquely human characteristic (Goodall, 1964). The chimpanzee study sites that exhibit termite fishing or consumption were outlined in the previous chapter (Table 4.1 ). The only site where chimpanzees have been studied and are known to consume termites but not use tools is

Budongo Forest. Here, the chimpanzees knock over mounds to eat Cubitermes (Newton

Fisher, 1999). Cubitermes are not known to be eaten by chimpanzees elsewhere. Such behavior at Budongo may be due to a low frequency of Macrotermes (Newton-Fisher, 1999), but these analyses have yet to be done. This case is isolated, and most chimpanzees consume

Macrotermes species and use tools to obtain them.

Fongoli is the one of two dry savanna habitat sites where chimpanzees are known to termite fish. It may be assumed that this site will produce different results compared to most other sites based on significant habitat and climatic differences. One difference is that the

Fongoli chimpanzees termite fish throughout the year (Pruetz, 2005). Therefore, many factors examined in this study were included in order to examine whether there is a difference between dry and wet season termite fishing, how tools compare to other sites, and what the selective pressures for this behavior at Fongoli may be. For example, McBeath and

McGrew (1982) found that termite fishing tools were most often made from plants growing within two meters of chimpanzee termite mounds and were often of the species Grewia lasiodiscus. It was therefore expected that the materials used for termite fishing tools by 48

Fongoli chimpanzees would come mainly from the Grewia, due to the similarities in habitat between these two sites.

Data Collection

Chimpanzee termite-fishing tools were collected when found and further analyzed regarding plant material, lengths, diameters, and characteristics of the proximal and distal ends. Tool lengths and characteristics were analyzed for comparison to other sites. The distance from the mound to the tool was measured. The tools found at these mound sites were classed as fresh (less than 24 hours old), recent (2-4 days old), or old (more than 4 days old) (McBeath & McGrew, 1982), based on color and the use of a reference collection by the author. Age of assemblage was also determined from assessing the archaeology of the site, using indicators such as associated feces freshness, and freshness of breaks on the plant material used as a tool (McGrew et al., 2003). Tools were made from six possible types of raw materials: twig, vine, bark, leaf-stalk, grass, and palm-frond (McGrew et al., 1979).

Species of plant was noted if known. Availability of vegetation around the mound was recorded as well as its distance from the mound, and percent coverage from vegetation above the mound was assessed. Seasonality was determined by rainfall: the dry season ended on

June 25th when rain became frequent (daily) and heavy (ground flooding and river flow); the wet season began on June 26th.

Results

Plant material and vegetation

Of the vegetation around fished termite mounds, a few species were prominent. Saba

senegalensis (22%), Grewia sp (19%), and Hexalobus monopetalus (15%) were the plants

most frequently encountered. However, there was no significant difference across species of 49 plants in frequency found in proximity to termite fished mounds (GLM procedure f=0.59, df= 19, p=0.9, two-tailed). Figure 5.1 shows a chimpanzee with a plant in his mouth, ready to modify it.

The raw material sources for tools (Figure 5.1) were all within 10 meters of the mounds. Grewia grew a mean distance of2.5 meters (range 0-lOm) from the mound, while

Hexalobus had a mean of0.7 m (range 0-Sm) and Saba 1.1 meters (range 0-10 m) from the mound.

The average percent foliage coverage for all mounds recorded is 41 %, with a range of

0-100%. Of the fished mounds, the average foliage cover is 63%, with a range of 15-100%.

Termite mounds selected for fishing were never completely in the open (i.e. without tree coverage).

Figure 5.1: Chimpanzee holding plant material in mouth. 50

Figure 5.2: Fresh tools found in situ at mound TM161 a.) overview of tools found b.) Tool made of Saba.

Figure 5.3: Tool kit of Saba at mound TM161 (3 flexible probes and 2 clippings).

Tool characteristics

Fishing probes were aged when collected; 401 of 533 (75%) were fresh or recent, and only these will be used for analysis. Figure 5.2 illustrates a fresh find of tools made from

Saba, and Figure 5.3 shows a tool kit for that particular find. Tools were most commonly made from Grewia, Hexalobus, and Saba plants (77%). Plant sources for tools were, on 51 average, 1.3 m from the termite mound (range 0-10 m). The average distance between tools and the mound is 36.5 cm with a range of 0-224 cm (Table 5.1 ). All tools were found within

2.5 meters of a mound. All 401 fresh tools were flexible probes, with a mean length of29.5 cm (range 9-106 cm) (Table 5.1) made of wood, twig, or vine materials, but not grass (Figure

5 .4a). Dry season tools had a mean length of 31.3 cm with a range of 11.9-106 cm. Wet season tools were characterized by a mean length of 26. 7 cm with a range of 9 .2-72 .3 cm.

However, there was no significant difference between tools from the different seasons

(Poisson analysis, p-value=0.5434, two-tailed). Frayed distal ends of tools occurred on 61 % of all flexible probes (Figure 5 .4b ), while the proximal ends were stripped or clipped (Figure

5.4c). The average diameters of the proximal and distal ends did not differ greatly (Table

5.1).

On June 30th of 2004, a perforating stick made from bamboo that was 20.5 cm in length with a 0.7 cm diameter was found next to a mound with 36 flexible probes (mound

TM152) (Figure 5.5 shows perforating stick). This is one of two perforating sticks found at

Fongoli (Pruetz, unpublished data).

Table 5.1: Flexible probe measurements (cm). Tool Dry Wet Proximal Distal I Distance I I lenQth season season I end I end I from mound Average I 29.5 31 .3 26.7 0.027 0.021 I 36.5 I I Max I 106 106 72.3 1 o.s T 224 I I I Min 9 11 .9 9.2 0.1 0.1 0 52

c.)

Figure 5.4: Tool characteristics a.) flexible probe of twig material from mound 124NN b.) frayed distal end of 124NN tool c.) stripped proximal end of Grewia tool at mound 103NN.

Figure 5.5: Perforating stick of Bamboo from mound TM152. 53

Discussion and Conclusions

Fongoli resembles Mt. Assirik in the types of material used by chimpanzees for termite fishing. At Fongoli, grass was not a material source in this study due to season. Grass was not tall enough at the end of the dry season and at the beginning of the wet season for tool material. McGrew and colleagues (1979) found twigs to be the prominent material of choice at Mt. Assirik in Senegal. At Assirik, 97% of tools made were of twigs, leaf-stalks, and vines (McBeath & McGrew, 1982), while at Gombe the majority of tools were made from grass (McGrew et al., 1979). At Mahale the majority of tools were made of bark (75%)

(McGrew & Collins, 1985). Ecology may be a factor contributing to these differences, but this possibility has not been discussed in detail regarding these sites.

Termite fishing tools were mostly made from three plant species typically found around mounds used for termite fishing. These were species that the chimpanzees also consumed (Pruetz, 2005). Grewia, Hexalobus, and Saba combined were most often selected for source material at Fongoli. However, this finding differs from data recorded at Mt.

Assirik, in which Grewia lasiodiscus (twig) alone was the species of plant that the chimpanzees used most often (McGrew et al., 1979; McBeath & McGrew, 1982).

Saba at Fongoli was found on average 1.1 meters (range 0-1 Om) from fished mounds.

McGrew and colleagues (1979) found that 94% of the tools originated from plants within two meters of the mounds at Mt. Assirik, Gombe, and Rio Muni. This study shows that tools can originate from sources up to 10 meters away, but on average the distance to source was only

1-2m. This suggests that chimpanzees have plentiful raw materials for probes at Fongoli, similar to what has been reported for other sites. 54

The Fongoli chimpanzees select termite mounds that are generally well shaded from direct sunlight. This could be a thermoregulatory factor for the chimpanzees, or it may be a by-product of the frequency with which the chimpanzees' use woodland versus closed habitats. Shade is not considered to be a selective pressure for termite fishing here, but is a secondary characteristic resulting from habitat selectivity.

Flexible probe length (all tools found) reported in this study did not differ significantly (Poisson t-test with 95% CI, p-value=0.27, two-tailed) from the preliminary study done at Fongoli (McGrew et al., 2005). However, grass material was used in the latter study, which was not found to be used for tools here. This discrepancy is due to seasonality.

In the dry and early wet seasons in which this research was done, the grass was not tall enough to make fishing probes. The range of tool lengths was also greater in this study, possibly reflecting the greater number of tools found versus the preliminary study done by

McGrew and colleagues (2005). There was no significant difference between tools from the dry season and the wet season (Poisson analysis, p-value=0.5434, two-tailed). However further investigation and greater sample sizes could prove otherwise; dry season tools exhibit a trend toward a longer range and average compared to the wet season. Data on dry season tools are limited and should be further researched. Shorter probes had been clipped and the ends were left as debitage at mounds. Tool lengths from sites are compared in Table 5.2.

Shorter probes may indicate easier access to termites.

The perforating stick found is anecdotal in this study due to the small sample size

(McGrew & Marchant, 1997), but the awareness that the Fongoli chimpanzees may use

. perforating sticks allows us to look for it in the future. These sticks are not used at every site characterized by termite fishing and they are hypothesized to be replacements for manual 55 removal of the termite soil to uncover holes. Some sites that use perforating sticks include

Ndoki, Goualougo, and the Ndakan and Bai Hokou sites in Congo (Suzuki et al., 1995; Sanz et al., 2004; Fay & Carroll, 1994).

Termite fishing was observed three times during this study. These observations gave evidence as to the technique used by the Fongoli chimpanzees (Pruetz, unpublished data).

The probe is inserted into the termite mound after opening a hole with the fingers, and after extraction the chimpanzees often support the probe on the opposite wrist and pick off the termites with their lips. Picking of termites between two fingers (not thumbs) off of mounds was observed. A male was observed to bring a tool from at least 6 meters away to a mound.

A total of 5 individuals were observed termite fishing (3 males, 2 females, 1 infant watching mother and attempting to fish with her discarded tool). Fongoli chimpanzees' technique resembles that described for Gombe and (Figure 5.7) (Goodall, 1963; 1968) Mahale,

Tanzania (McGrew & Collins, 1985), and Goualougo, Congo Basin (Sanz et al., 2004).

Table 5.2: Flexible probe lengths of chimpanzee sites. s·Ite [ T 00I I en gth 1 s ource

I mean (cm) Fongoli 29.5 This study Mt. Assirik 32.5 McGrew et al., 1979

Gombe 30.7 McGrew et al., 1979

-----·~----~--· --- Mahale 37.7 McGrew & Collins, 1985

Rio Muni 49.7 McGrew et al., 1979

Ndoki 50.8 Suzuki et al., 1995

I ----· -···-·- Goualougo 43.1 Sanz et al., 2004 I 56

Figure 5.6: Gombe chimpanzee termite fishing. 57

CHAPTER 6: Conclusions and Summary

Termite Fishing Behavior of Fongoli Chimpanzees

Fongoli chimpanzees consume insects mainly in woodland habitats. However, they also termite fish in the ecotone and gallery forest habitats significantly more than would be expected based on availability of these sites. These findings contrast with Assirik's chimpanzees termite fishing mainly in ecotone habitats (McBeath & McGrew, 1982). This could be because Assirik has more ecotone (5%) (McBeath & McGrew, 1982) compared to

Fongoli (3%). Assirik's ecotone is characterized by a clear transition between habitats and is easy to see, whereas at Fongoli it is more subtle (Pruetz, personal communication). This may be a factor in the selection of ecotone at Fongoli as well. The available vegetation may also be a factor. Fongoli chimpanzees select three plant species predominantly for making tools over all others, while chimpanzees at Assirik select one species of plant material over all others (McBeath & McGrew, 1982). This selection may be due to the fact that these plants are abundant around the mounds.

Termite species play an important role in determining termite fishing behavior of

Fongoli chimpanzees. Two genera and four species of termites were found in association with tools at Fongoli. Such diversity is greater than that associated with tools at any other chimpanzee site. However, direct evidence for chimpanzee consumption of Trinervitermes at

Fongoli has not yet been found. Several explanations could account for this. First, the

Fongoli chimpanzees do not fish for Trinervitermes. These termites are only sharing mounds with Macrotermes, which have been found through direct evidence to be eaten (observation and feces). Another explanation may be that Fongoli chimpanzees do not termite fish for

Trinervitermes but do consume them by picking them up with their hands. Trinervitermes are 58 known to forage on the surface and are also more active in the dry season (Ohiagu, 1979).

This species may therefore be a replacement for Macrotermes, which are more difficult to obtain during this time (Noirot & Darlington, 2000). Finally, both Trinervitermes and

Macrotermes may be consumed. Possible explanations for the lack of evidence for the consumption of Trinervitermes are that the heads of this genus are not indigestible like those of Macrotermes or that it is difficult to discern the smaller heads of Trinervitermes through macro-analyses of feces. Further research could be conducted to examine this possibility.

The Fongoli termite fishing tool kit does not differ significantly from that of most other termite fishing chimpanzee sites. Based on direct observation, the technique used by

Fongoli chimpanzees is similar to Gombe (Goodall, 1963; 1968) and Mahale, Tanzania

(McGrew & Collins, 1985), and Goualougo, Congo Basin (Sanz et al., 2004). However, the dry season tools do exhibit a trend towards a longer length than those of the wet season. Such a trend may be due to Macrotermes' habit of retreating further underground in the dry season

(Noirot & Darlington, 2000). Another factor may be terrestriality. At Bossou, Guinea, chimpanzees were observed to spend more time on the ground during the dry season and more time arboreal in the wet season, which was attributed to a reduction of thermoregulation costs by resting in trees during the cooler months (Takemoto, 2004). Since the treetops would have a higher concentration of heat from the sun than the forest floor in the dry season, terrestriality is increased (Takemoto, 2004). If this were true for Fongoli, the chimpanzees may consume termites in the dry season due in part to their increased terrestrial activity.

Termite mound density does not seem to influence termite fishing at Fongoli, but rather the habitat is an important influence on this behavior in chimpanzees. Vegetation 59 surrounding fished mounds was proportionately more shaded from direct sunlight, which may also be a by-product factor of the habitat influencing chimpanzee termite fishing selection. Macrotermes are more prominent in woodland habitats and this may also be a selecting factor for the Fongoli chimpanzees. Therefore, many selective pressures influence the Fongoli chimpanzees as to where and what they termite fish.

The Fongoli chimpanzee diet consists of 5.4% invertebrates (Pruetz, 2005). The consumption of insects does not increase during times of fruit scarcity (Pruetz, 2005). Thus, termites at Fongoli are not considered a fallback food. Further research would provide evidence as to whether termites are a staple in the diet.

Benefits from Eating Termites

In terms of nutrition, Wrangham (1977) found that Macrotermes workers and immatures provide the greatest amount of protein (68.4%), while reproductive individuals provided the highest percentage oflipids (52.8%); soldiers fell between the two castes on both accounts (Figure 6.1 ). Termites are high in protein, rich in fat, and provide a high caloric intake (Berenbaum, 1995). Insects are especially important in compensating for dietary deficiencies. Many of the nutrients found in termites are used by African peoples who are malnourished (Berenbaum, 1995). Table 6.1 displays the various nutrients and their content for a termite species compared to beef, chicken and fish (taken from Berenbaum,

1995: p.179, Table 6.1). A lack of iron can produce anemia (growth and blood loss require the nutrient iron). The nutrient thiamine helps the nervous system, synthesis of collagen, improving cognitive function, and maintaining memory, while also playing a role in promoting growth and repair of all body tissues. Riboflavin is used in body growth and red blood cell production and helps in releasing energy from carbohydrates. Niacin reduces the 60 amount of cholesterol and certain fatty substances in ·blood. Iron, thiamine, riboflavin, and niacin are all used to aid an inadequate diet in humans (Berenbaum, 1995), and would thus also provide chimpanzees nutrients important for growth and maintenance of health

(Wrangham, 1980).

Insects are eaten throughout most of the world by human populations living in Africa,

Asia, and Latin America (DeFoliart, 1999). It has been suggested that Western populations

(Europeans and North Americans) do not consume insects due to the unpredictability of supply and a constant supply from other sources (DeFoliart, 1999). Insects have been an important food throughout human history. In ancient human societies, such as those of the

Greeks and Romans, insects were cooked and considered a delicacy, prominent among upper class society (DeFoliart, 1999).

Figure 6.1: Macrotermes sp. soldier and immatures. 61

Table 6.1: Protein, vitamins, and nutrients in termites, beef, chicken, and fish (11 OOg ). Nutrient Energy Protein Calcium Phosphorus Iron Thiamine Niacin Riboflavin (kcal) (g) (g) (mg) (mg) (mg) (mg) (mg) Macrotermes 613 14.2 0.04 0.438 7.5 0.13 9.5 1.15 subhyalinus ! 0.09 Beef 219 27.4 0.012 0.23 3.5 - 6 I 0.23 Chicken 166 31.6 I 0.011 0.265 1.3 0.04 10.7 - -·- I 0.1 -- Fish 170 28.5 0.032 0.274 1 0.08 3 : 0.11

Termites diversified in the Cretaceous along with angiosperms (Thome et al., 2000).

The diversification of angiosperms has been proposed to correlate with primate origins by

Sussman (1987) while visual predation on insects was proposed by Cartmill (1972).

Angiosperm diversification and insect predation are considered to be evolutionary explanations for primate specializations in vision (color, stereoscopic vision) and grasping abilities. Social insects, such as termites, may have been important in the diet of primates throughout this lineage's existence.

Figure 6.2: Macrotermes sp. workers and soldiers. 62

Macrotermes

Macrotermes (Figure 6.2) are consumed at most sites by chimpanzees (see Table 4.1).

Macrotermes is the largest social insect in Africa and mounds contain an impressive number of individuals (Howse, 1970). One mound can contain up to 2 million Macrotermes (East

Africa: Harris, 1955; West Africa: Luscher, 1955). During the dry season, termites retreat far underground to escape the heat of the sun because of susceptibility to water loss (Noirot &

Darlington, 2000). Soil moisture affects the activity of all termite species; heavy rainfall will often cause termites to retreat further underground as well (Ferrar, 1982). Macrotermes have been known to descend to depths of greater than 40 m (Lepage et al., 197 4 ). Macrotermes ' underground passages are complex, consisting of three different kinds: radial, big cross and small cross passages (Jmhasley &Leuthold, 1999). These passages can extend up to 30 meters from the mound, and will often cross with those of other termite mound passages

(Jmhasley &Leuthold, 1999).

Hominid Insectivory?

Termite fishing in savanna chimpanzees is especially relevant to the study of early hominid behavior and ecology. Termites have recently been inferred as included in the diet of Australopithecus robustus, based on evidence from bone tools thought to be used to extract termites from mounds (Backwell & d'Errico, 2001). According to dental traits, gracile australopithecines are said to have had a diet much like modem apes today, i.e. fairly omnivorous (Teaford & Ungar, 2000). It is most likely that hominids and even protohominids used tools before such evidence appears in the fossil record at about 2.6 mya

(Beck, 1980; McGrew, 1992; Sigaut, 1993). Tool use by early hominids has been hypothesized as an important development in the evolution of human diet (Ungar & Teaford, 63

2002). Tools used by nonhuman primates for food procurement that were made of organic materials would not survive fossilization (Teaford et al., 2002). Evidence of these tools in the archaeological record is absent due to decay of organic matter such as, wood, twigs, sticks, and leaves. It is conceivable that early hominid tools were similar to what we find in extant nonhuman primates (Teaford et al., 2002). Since tool use is found in great apes, it is not unreasonable to assume that the last common ancestor may have had some simple form of tool technology (about 6 million years ago). Stone and bone tools of early hominids may have been used similar to those of nonhuman primates; however functional interpretation is difficult (Teaford et al., 2002). Could such tool use have led to the evolution of stone tools?

Stone tools have been recorded in the Tai community of chimpanzees for the cracking of nuts

(Boesch et al. 1994), and attempts have been made to relate this behavior to early hominid stone tool techniques (Joulian, 1996; Panger et al., 2002). Early hominids may have used similar methods of extracting invertebrates, cracking nuts, and other food related tool use like that observed in non-human apes today (Foley & Lahr, 2003; Panger et al., 2002). Panger and colleagues (2002) suggest that, while direct evidence for stone tools dates to 2.6 million years ago, Australopithecus may have had the manipulative capabilities for tool making and using around 3 .2 million years ago. Systematic study concerning the prehistoric diet, especially regarding insects and nuts, may allow more direct inferences for simple forms of technology made from materials like that used by living non-human apes. Such materials may have been used for termite fishing, ant dipping and fishing, nut cracking, and honey extraction. Insects in the early hominid diet could have provided protein and nutrients that plants and fruits could not provide, especially if hunting and scavenging were limited. Early hominids may have been more reliant on insects than large vertebrates. Later hominids might 64 have been characterized by a decrease in invertebrates and an increase in vertebrate prey, which would also be significant in the evolution from simple forms of tools to complex tools.

Some believe that a significant change in human diet occurred between Australopithecus and

Homo, with the latter having a dramatic increase in meat procurement (Milton, 1999).

Perishable tools may have assisted early Homo in food processing and procurement, allowing a variable diet (Teaford et al., 2002). As tooth size and enamel thickness decreased, the use of tools for food processing likely became even more important (Teaford et al., 2002).

Isotopic analyses done on Australopithecus africanus and robustus demonstrate that early hominids exploited open habitats for food, such as woodland and grassland

(Sponheimer & Lee-Thorp, 1999). This conclusion may also lead to further investigation of whether early hominids consumed a higher intake of animal prey than once believed

(Sponheimer & Lee-Thorp, 1999). It may be that open habitats do not provide as much fruit as do more forested ones. Fongoli's mosaic habitat and the chimpanzees that live there will help in providing referential models for the Paleolithic diet of early hominids.

The impact ofFongoli chimpanzees' emphasis on invertebrates may be interpreted regarding the early hominid diet in terms of an important resource that has been ignored quite frequently by researchers (see Ungar & Teaford, 2002). Fongoli is only one of a few dry savanna habitats where chimpanzees have been studied and provides evidence for the chimpanzees' reliance on termites throughout the year. Fongoli and Assirik have environments similar to those that have been proposed for Plio-Pleistocene hominids in East

Africa (McGrew et al., 1981). Considering the environment at Fongoli, chimpanzees here may provide insight into early hominid life. Consuming insects throughout the year rather 65 than as a fallback food may also characterize the diet of early hominids. This study suggests that insects may have been a crucial element in early hominid diet.

Questions of Culture

Whiten and colleagues (2001) have mapped out what they define as cultural variation among chimpanzees from different sites across Africa. As of 2001 , a total of nine sites are considered. A recent conference was held in order to add more sites, which will include

Fongoli (Pruetz, personal communication). This charting of cultural behaviors illustrates the differences between communities of chimpanzees. Chimpanzee site directors are given a list of behaviors to record for at their respective sites (Whiten, 2001; Whiten et al., 2001 ). These behaviors are then assigned patterns of representation: not known, not possible, absent, present, habitual, and customary (Whiten, 2001; Whiten et al., 2001). There are a total of 39

'cultural' behaviors recorded so far (Whiten et al., 2001), and more may be added in the next attempt. From this effort, primatologists can now compare behaviors found at their chimpanzee sites to those of others. This attempt is similar to what anthropologists have been doing with modem humans for centuries. Since this initial effort to chart culture in non humans, there are now moves to do so among dolphins, whales (Whitehead et al., 2004), capuchins (Fragaszy et al., 2004) and orangutans (van Schaik et al., 2003b).

The identification of chimpanzee culture helps to illustrate their welfare and significance to humans who may not be aware of the chimpanzee's impeding threat, with the ultimate goal of conservation in mind. Today, the greatest threat to primate populations is habitat destruction (Hill, 2002). According to a recent survey, as of2003, the western subspecies of chimpanzee is now extinct in at least two countries in which they once resided:

Benin and Togo (Butynski, 2003). Estimates in 2003 estimate a population of200,000 for all 66 subspecies of Pan troglodytes combined, and all species and subspecies of Pan are considered "endangered" according to the United States Endangered Species Act of 1973

(Butynski, 2003). Nigerian (Pan troglodytes vellerosus) and Western chimpanzees (Pan troglodytes verus) (Figure 6.3) have the lowest population numbers and are currently at the greatest risk (Butynski, 2003). Senegal is reported to have between 100 and 300 Western chimpanzees left in the wild (Carter et al., 2003). Thus, research on these chimpanzees is greatly needed before they disappear. Chimpanzees live a tenuous existence in the

"marginal" habitats of Fongoli (Pruetz, unpublished data). Research on diet in chimpanzees is beneficial in terms of conservation of habitat. This research shows the importance of closed habitats (ecotone and gallery forests) as well as open habitats (woodland) in the consumption of termites, which is also a major contributor to the Fongoli chimpanzee diet.

Thus, conservation of these habitats is important for the survival and diet of the chimpanzees.

Many questions addressed here have yet to be answered but are on-going questions that drive anthropologists to conduct research. We may never understand completely how human culture evolved. However, we are aware that our own culture may eventually destroy us. Looking for answers to how and why human culture evolved may provide answers that help us stop the destruction. Anthropologists using paleo-studies and extant primate studies may hold keys to the answers. We can make inferences on cultures from the past and from non-human primates, and we can only hope that we get close enough to the truth with our knowledge to get a better picture of human evolution. With such knowledge it may be possible to understand where the human species is headed. 67

Figure 6.3: Adolescent male, West African chimpanzee (Pan troglodytes verus).

Much debate has been on whether non-human animals can have culture (Foley &

Lahr, 2003; Laland & Hoppitt, 2003). As more information is compiled it will be difficult to deny it any longer, and the human species will again theoretically decrease regarding our superiority that we believe we hold in this universe. It is a fact that people are not ready to accept, but it may be too late when they do, for all creatures and environments that have fascinated us may be gone by then.

Future Research

I will continue my research into the insectivorous diet of chimpanzees in search for answers to hominid evolution and chimpanzee habitat conservation. Forested habitats are extremely important for chimpanzee diet, as I have illustrated here and others have shown regarding the frugivorous diet of the chimpanzee (Pruetz, 2005; Wrangham et al., 1993;

Sugiyama & Koman, 1992). 68

F ongoli is a dry mosaic habitat that is not often found associated with chimpanzees.

Another site, Semliki, Uganda is similar to Fongoli's habitat, but the chimpanzees have not shown any evidence for the consumption of invertebrates such as termites (Hunt & McGrew,

2002). Research into this area has not been completed. A possible explanation may be that the knowledge of termite fishing is absent; however there are ways of obtaining termites and other invertebrates without the use of tools (Newton-Fisher, 1999; McGrew, 1992), which should be examined further.

I propose a more extensive methodology examining termite diversity across sites to give a better understanding of the relation between termite availability and consumption.

Termite mound density and species composition at other termite fishing chimpanzee sites would give the necessary tools for comparison across sites to determine variations. Micro analysis of termite remains in the feces may help to give more detailed results of chimpanzee invertebrate diet. Further observational data on Fongoli chimpanzee termite fishing may give us a better understanding of the diversity of termite species in the diet and the importance of termites in the diet overall. 69

REFERENCES CITED

Balcomb SR, Chapman CA, Wrangham RW. 2000. Relationship between chimpanzees (Pan troglodytes) density and large, fleshy-fruit tree density: conservation implications. American Journal ofPrimatology 51:197-203.

Baldwin PJ. 1979. The Natural History ofthe Chimpanzee (Pan troglodytes verus), at Mt. Assirik, Senegal. Ph.D. Thesis: University of Sterling

Baldwin PJ, Sabater-Pi J, McGrew WC, Tutin CEG. 1982. Comparisons of nests made by different populations of chimpanzees (Pan troglodytes). Primates 22:367-385.

Basabose AK. 2002. Diet composition of chimpanzees inhabiting the montane forest of Kahuzi, Democratic Republic of Congo. American Journal ofPrimatology 58: 1-21.

Beck BB. 1980. Animal Tool Behavior: The Use and Manufacture of Tools by Animals. New York: Garland Publishing.

Berenbaum MR. 1995. Bugs in the System: Insects and Their Impact on Human Affairs. Reading, Massachusetts: Addison-Wesley Publishing Company.

Backwell LR, d'Errico F. 2001. Evidence of termite foraging by Swartkrans early hominids. Proceedings of the National Academy ofSciences of the United States ofAmerica 98:1358- 1363.

Balcomb SR, Chapman CA, Wrangham RW. 2000. Relationship between chimpanzee (Pan troglodytes) density and large, fleshy fruit tree density: conservation implications. American Journal ofPrimatology 51: 197-203.

Boesch C. 2003. Is culture the golden barrier between human and chimpanzee? Evolutionary Anthropology 12:82-91.

Boesch C, Boesch H. 1990. Tool use and tool making in wild chimpanzees. Folia Primatologica 54: 1-15.

Boesch C, Marchesi P, Marchesi N, Fruth B, Joulian F. 1994. Is nut cracking in wild chimpanzees a cultural behaviour? Journal ofHuman Evolution 26:325-338.

Bohannan P, Bohannan L. 1968. Tiv Economy. Evanston: Northwestern University Press.

Boinski S, Quatrone R, Swartz H. 2000. Substrate and tool use by brown capuchins in Suriname: ecological contexts and cognitive bases. American Anthropology 102:741-761.

Budenheimer FS. 1951. Insects as Human Food. Netherlands: Dr. W. Junk, The Hague. 70

Butynski TM. 2003. The robust chimpanzee (Pan troglodytes): taxonomy, distribution, abundance and conservation status. In Kromos R, Boesch C, Bakarr MI, Butynski TM, editors, West African Chimpanzees: Status Survey and Conservation Plan. Gland, Switzerland and Cambridge, UK: Union for Conservation of Nature and Natural Resources, p. 5-12.

Byrne RW, Corp N, Byrne JME. 2001. Estimating the complexity of animal behavior: how mountain gorillas eat thistles. Behaviour 138:525-557.

· Carter J, Ndiaye S, Pruetz JD, McGrew WC. 2003. Senegal. In: Kormos R, Boesch C, Bakarr MI, Butynski TM, editors. West African Chimpanzees: Status Survey and Conservation Action Plan. IUCN/SSC Primate Specialist Group, p. 31-39

Cartmill M. 1972. Arboreal adaptations and the origin of the order Primates. In: Tuttle R, editor. The Functional and Evolutionary Biology ofPrimates. Chicago: Aldinc, p. 97-122.

Collins DA, McGrew WC. 1985. Chimpanzees' (Pan troglodytes) choice of prey among termites (Macrotermitinae) in western Tanzania. Primates 26:375-389.

Collins DA, McGrew WC. 1987. Termite fauna related to differences in tool-use between groups of chimpanzees (Pan troglodytes). Primates 28: 457-471.

Collins DA, McGrew WC. 1988. Habitats of three groups of chimpanzees (Pan troglodytes) in western Tanzania compared. Journal ofHuman Evolution 17:553-574.

Cottam G, Curtis JT. 1956. The use of distance measures in phytosociological sampling. Ecology 37:451-461.

DeF oliart GR. 1999. Insects as food: why the western attitude is important. Annual Review of Entomology 44:21-50.

Eggleton P, Bignell DE, Hauser S, Dibog L, Norgrove L, Madong B. 2002. Termite diversity across an anthropogenic disturbance gradient in the humid forest zone of West Africa. Agriculture, Ecosystems and Environment 90: 189-202.

Fay JM, Carroll RW. 1994. Chimpanzee tool use for honey and termite extraction in Central Africa. American Journal ofPrimatology 34:309-317.

Ferrar P. 1982. Termites of the South African savanna: IV. subterranean populations, mass determinations and biomass estimations. Oecologia 52: 14 7-151.

Foley R, Lahr MM. 2003. On stony ground: lithic technology, human evolution, and the emergence of culture. Evolutionary Anthropology 12: 109-122. 71

Fossey D, Harcourt AH. 1977. Feeding ecology of free-ranging Mountain gorilla (Gorilla gorilla beringei). In: Clutton-Brock TH, editor. Primate Ecology: Studies offeeding and ranging behaviour in lemurs, monkeys and apes, London, New York: Academic Press p. 415-447.

Fox EA, Muhammad lb. 2002. Brief communication: new tool use by wild Sumatran orangutans (Pongo pygmaeus abelii). American Journal ofPhysical Anthropology 119: 186- 188.

Fox EA, van Schaik CP, Sitompul A, Wright DN. 2004. Intra-and interpopulational differences in orangutan (Pongo pygmaeus) activity and diet: implications for the invention of tool use. American Journal ofPhysical Anthropology 125:162-174.

Fragaszy D, Izar P, Visalberghi E, Ottoni EB, de Oliveira MG. 2004. Wild capuchin monkeys (Cebus libidinosus) use anvils and stone pounding tools. American Journal of Primatology 64:359-366.

Fruth B, Gottfried H, McGrew WC. 1999. The Pan species. In: The Nonhuman Primates Phyllis Dolhinow and Agustin Fuentes eds. Boston: McGraw-Hill p. 64-72.

Goodall J. 1963. Feeding behaviour of wild chimpanzees. Symposia of the Zoological Society ofLondon 10:39-48.

Goodall J. 1964. Tool-using and aimed throwing in a community of free-living chimpanzees. Nature 201:1264-1266.

Goodall JvL. 1968. The behaviour of free-living chimpanzees in the Gombe Stream Reserve. Animal Behaviour Monographs 1:161-311.

Hart BL, Hart LA, McCoy M, Sarath CR. 2001. Cognitive behavior in Asian elephants: use and modification of branches for fly switching. Animal Behavior 62:839-847.

Hill CM. 2002. Primate conservation and local communities- ethical issues and debates. American Anthropologist 104: 1184-1194.

Hladik CM. 1977. Chimpanzees of Gabon and chimpanzees of Gombe: some comparative data on the diet. In: Clutton-Brock TH, editor. Primate Ecology: Studies offeeding and ranging behaviour in lemurs, monkeys and apes, London, New York: Academic Press p. 481-501.

Howse PE. 1970. Termites: A Study in Social Behaviour. London: Hutchinson University Library. 72

Humle T, Matsuzawa T. 2001. Behavioral diversity among the wild chimpanzee populations ofBossou and neighboring areas, Guinea and Cote d'Ivoire, West Africa. Folia Primatologica 72:57-68.

Hunt KD, McGrew WC. 2002. Chimpanzees in the dry habitats of Assirik, Senegal and Semliki Wildlife Reserve, Uganda. In: Boesch C, Hohmann G, Marchant LF, editors. Behavioural Diversity in Chimpanzees and Bonobos, Cambridge: Cambridge University Press p. 35-51.

Isbell LA. 1998. Diet for a small primate: insectivory and gummivory in the (large) patas monkey (Erythrocebus patas pyrrhonotus ). American Journal ofPrimatology 45 :3 81-398.

Jrnhasly P, Leuthold RH. 1999. The system of underground passages in Macrotermes subhyalinus and comparison oflaboratory bioassays to field evidence of intraspecific encounters in M subhyalinus and M bellicosus (Isoptera, Termitidae). Insectes Sociaux 46:332-340.

Jolly A. 1985. The Evolution ofPrimate Behavior. New York: Macmillan Publishing Company.

Joulian F. 1996. Comparing chimpanzee and early hominid techniques: some contributions to cultural and cognitive questions. In: Mellars PA, Gibson KR, editors. Modeling the Early Human Mind, Cambridge: Cambridge University Press p. 173-188.

Kawai M. 1965. Newly-acquired pre-cultural behavior of the natural troop of Japanese monkey on Koshima islet. Primates 6:1-30.

Kenwood B, Weir AAS, Rutz C, Kacelnik A. 2005. Tool manufacture by naive juvenile crows. Nature 433:121.

Krob J, Linsenmair KE. 2001. The causes of spatial patterning of mounds of a fungus cultivating termite: results from nearest-neighbour analysis and ecological studies. Oecologia 127:324-333.

Laland KN, Hoppitt W. 2003. Do animals have culture? Evolutionary Anthropology 12:150- 159.

Lee KE, Wood TG. 1971. Termites and Soils. London and New York: Academic Press.

Lepage M, Morel G, Resplendino C. 1974. Discovery of termite galleries extending deep underground to water level in North Senegal. Computes rendus Hebdomadaires des Seances de L 'Academie des Sciences (serie d) 278:1855-1858.

Luscher M. 1955. Air circulation in Macrotermes mounds. Acta Tropica 12:289-307. 73

Matsumoto-Oda A, Hayashi Y. 1997. Nutritional aspects of fruit choice by chimpanzees. Folia Primatologica 70:154-162.

McBeath NM, McGrew WC. 1982. Tools used by wild chimpanzees to obtain termites at Mt. Assirik, Senegal: the influence of habitat. Journal ofHuman Evolution 11 :65-72.

McGrew WC. 1983. Animal foods in the diets of wild chimpanzees (Pan troglodytes): why cross-cultural variation? Journal ofEthology 1:46-61.

McGrew WC. 1992. Chimpanzee Material Culture: Implications for Human Evolution. Cambridge: Cambridge University Press.

McGrew WC. 2001. The other faunivory: primate insectivory and early human diet. In: Stanford CB, Bunn HT, editors. Meat-Eating and Human Evolution, Oxford: Oxford University Press p. 160-178.

McGrew WC. 2004. The Cultured Ape: Reflections on Cultural Primatology. Cambridge: Cambridge University Press.

McGrew WC, Tutin CEG, Baldwin PJ. 1979. Chimpanzees, tools, and termites: cross cultural comparisons of Senegal, Tanzania, and Rio Muni. Man 14:185-214.

McGrew WC, Baldwin PJ, Tutin CEG. 1981. Chimpanzees in a hot, dry and open habitat: Mt. Assirik, Senegal, West Africa. Journal ofHuman Evolution 10:227-244.

McGrew WC, Collins DA. 1985. Tool use by wild chimpanzees (Pan troglodytes) to obtain termites (Macrotermes herus) in the Mahale Mountains, Tanzania. American Journal of Primatology 9:47-62.

McGrew WC, Marchant LF. 1997. Using Tools as hand: manual laterality and elementary technology in Cebus spp. and Pan spp. International Journal ofPrimatology 18:787-810.

McGrew WC, Baldwin PJ, Marchant LF, Pruetz JD, Scott SE, Tutin CEG. 2003. Ethoarcheology and elementary technology of unhabituated wild chimpanzees as Assirik, Senegal, West Africa. PaleoAnthropology 1 :1-20.

McGrew WC, Pruetz JD, Fulton BJ. 2005. Chimpanzees use tools to harvest social insects at Fongoli, Senegal. Folia Primatologica 871:1-5.

Milton K. 1999. A hypothesis to explain the role of meat-eating in human evolution. Evolutionary Anthropology 8:11-21.

Moore J 1996. Savanna chimpanzees, referential models and the last common ancestor. In: McGrew WC, Marchant LF, Nishida T, editors. Great Ape Societies, Cambridge: Cambridge University Press p. 275-292. 74

Moura ACA, Lee PC. 2004. Capuchin stone tool use in Caatinga dry forest. Science 306: 1909.

Nash LT. 1984. Diet and breeding seasonality among prosimians. American Journal of Physical Anthropology 63: 198.

Navarro A, Barton NH. 2003. Chromosomal speciation and molecular divergence: accelerated evolution in rearranged chromosomes. Science 300:321-324.

Newton-Fisher NE. 1999. Termite eating and food sharing by male chimpanzees in the Budongo Forest, Uganda. African Journal ofEcology 37:369-371.

Nishida T, Uehara S. 1980. Chimpanzees, tools, and termites: another example from Tanzania. Current Anthropology 21 :671-672.

Noirot C, Darrlington JPEC. 2000. Termite nests: architecture, regulation and defense. In: Abe T, Bignell DE, Higashi M, editors. Termites: Evolution, Sociality, Symbioses, Ecology, Dordrecht, Boston, London: Kluwer Academic Publishers p. 121-140.

Norris S. 2002. Creatures of culture? Making the case for cultural systems in whales and dolphins. BioScience 52:9-14.

Ohiagu CE. 1979. A Quantitative study of seasonal foraging by grass harvesting termite, Trinervitermes geminatus (Wasmann), (Isoptera, Nasutitermitinae) in southern Guinea Savanna, Mokwa, Nigeria. Oecologia 40: 179-188.

Ottoni EB, Mannu M. 2001. Semi-free ranging tufted capuchin monkeys (Cebus paella) spontaneously use tools to crack open nuts. International Journal ofPrimatology 22:34 7- 358.

Panger MA. 1998. Object-use in :free-ranging white-faced capuchins (Cebus capucinus) in Costa Rica. American Journal ofPhysical Anthropology 106:311-321.

Panger MA, Brooks AS, Richmond BG, Wood B. 2002. Older than the Oldowan? Rethinking the emergence ofhominin tool use. Evolutionary Anthropology 11 :235-245.

Paterson JD. 2001. Primate Behavior: An Exercise Workbook. Illinois: Waveland Press, Inc.

Pearce MJ. 1997. Termites: Biology and Pest Management. New York: CAB International.

Phillips KA. 1998. Tool use in wild capuchin monkeys (Cebus albifrons trinitatis). American Journal ofPrimatology 46:259-261. 75

Plummer TW, Bishop LC, Ditchfield P, Hicks J. 1999. Research on late Pliocene Oldowan sites at Kanjera South, Kenya. Journal ofHuman Evolution 36: 151-170.

Pruetz JD, Marchant LF, Amo J, McGrew WC. 2002. Survey of savanna chimpanzees (Pan troglodytes verus) in southeastern Senegal. American Journal ofPrimatology 58:35-43.

Pruetz JD. 2005. Feeding ecology of savanna chimpanzees (Pan troglodytes verus) at Fongoli, Senegal. In: Boesch C, Hohmann G, Robbins MM, editors. Feeding Ecology in Apes and Other Primates, Cambridge: Cambridge University Press, in press.

Remis MJ. 1997. Western Lowland gorilla (Gorilla gorilla gorilla) as seasonal frugivores: use of variable resources. American Journal ofPrimatology 43:87-109.

Rendell L, Whitehead H. 2001. Culture in whales and dolphins. Behavioral and Brian Sciences 24:309-324.

Richard A. 1985. Table 5.1 In: Primates in Nature, New York: Freeman Press p. 164-165.

Rightmire GP. 2004. Brain size and encephalization in early to mid-Pleistocene Homo. American Journal ofPhysical Anthropology 124:109-123.

Ritchie BG, Fragaszy DM. 1988. Capuchin monkey (Cebus apella) grooms her infant's wound with tools. American Journal ofPrimatology 16:345-348.

Rodman PS. 1977. Feeding behaviour of orang-utans of the Kutai Nature Reserve, east Kalimantan. In: Clutton-Brock TH, editor. Primate Ecology: Studies offeeding and ranging behaviour in lemurs, monkeys and apes, London, New York: Academic Press p. 384-413.

Sanz C, Morgan D, Gulick S. 2004. New insights into chimpanzees, tools, and termites from the Congo Basin. American Naturalist 164:567-581.

Sarr M, Agbogba C, Russell-Smith A. 1998. The effects of length of fallow and cultivation on termite abundance and diversity in the sahelian zone of Senegal: a preliminary note. Pedobiologia 42:56-62.

Sigaut F. 1993. How can we analyse and describe technical actions. In: Berthelet A, Chavaillon J, editors. The Use of Tools by Human and Non-Human Primates, Oxford: Claredon Press p. 381-397.

Sponheimer M, Lee-Thorp JA. 1999. Isotopic evidence for the diet of an early hominid, Australopithecus africanus. Science 283:368-370.

Stanford CB, Bunn HT. 2001. Meat-Eating and Human Evolution. Oxford: Oxford University Press. 76

Sugiyama Y, Koman J. 1987. A preliminary list of chimpanzees' alimentation at Bossou Guinea. Primates 28:133-147.

Sussman RW. 1987. Species-specific dietary patterns in primates and human dietary adaptations. In: Kinzey WG editor. The Evolution ofHuman Behavior: Primate Models, Albany: State University of New York Press p. 143-156.

Suzuki S, Kuroda S, Nishihara T. 1995. Tool-set for termite-fishing by chimpanzees in the Ndoki Forest, Congo. Behaviour 132:219-235.

Takemoto H. 2004. Seasonal change in terrestriality of chimpanzees in relation to microclimate in the tropical forest. American Journal ofPhysical Anthropology 124:81-92.

Teaford MF, Ungar PS. 2000. Diet and the evolution of the earliest human ancestors. Proceedings ofthe National Academy ofScience 97:13506-13511.

Teaford MF, Ungar PS, Grine FE. 2002. Paleontological evidence for the diets of African Plio-Pleistocene hominins with special reference to early Homo. Ungar PS, Teaford MF, editors. Human Diet: Its Origin and Evolution, Westport, Connecticut, London: Bergin & Garvey p. 143-166.

Thome BL, Grimaldi DA, Krishna K. 2000.Early fossil history of the termites. In: Abe T, Bignell DE, Higashi M, editors. Termites: Evolution, Sociality, Symbioses, Ecology. Dordrecht, Boston, London: Kluwer Academic Publishers p. 77-94.

Tutin CEG, McGrew WC, Baldwin PJ. 1983. Social organization of savanna-dwelling chimpanzees, Pan troglodytes verus, at Mt. Assirik, Senegal. Primates 24: 154-173.

Tutin CEG, Ham RM, White LJT, Harrison MJS. 1997 The primate community of the Lope Reserve. Gabon: diets, responses to fruit scarcity, and effects on biomass. American Journal ofPrimatology 42:1-24.

Turnbull CM. 1965. Wayward Servants: The Two Worlds ofthe African Pygmies. Garden City, New York: The Natural History Press.

Uehara S. 1982. Seasonal changes in the techniques employed by wild chimpanzees in the Mahale Mountains, Tanzania, to feed on termites (Psuedocanthotermes spininger). Folia Primatologica 37:44-76.

Ungar PS, Teaford MF. 2002. Human Diet: Its Origin and Evolution, Westport, Connecticut, London: Bergin & Garvey.

van Schaik CP, Deaner RO, Merrill MY. 1999. The conditions for tool use in primates: implications for the evolution of material culture. Journal ofHuman Evolution 36:719-741. 77 van Schaik CP, Fox EA, Fechtman LT. 2003a. Individual variation in the rate of use of tree-hole tools among wild orang-utans: implications for hominine evolution. Journal of Human Evolution 44:11-23. van Schaik CP, Ancrenaz M, Borgen G, Galdikas B, Knott CD, Singleton I, Suzuki A, Utami SS, Merrill M. 2003 b. Orangutan cultures and the evolution of material culture. Science 299: 102-105.

Visalberghi E, Fragaszy D. 1990. Do monkeys ape? In: Parker ST, Gibson KR, editors. Language and Intelligence in Monkeys and Apes, Cambridge: Cambridge University Press p. 247-273.

Visalberghi E, McGrew WC. 1997. Cebus meets Pan. International Journal ofPrimatology 18:677-681. de Waal F. 2001. The Ape and the Sushi Master, New York: Basic Books.

Warren RD, Crompton RH. 1998. Diet, body size, and the energy costs oflocomotion in saltatory primates. Folia Primatologica 69:86-100.

Whitehead H, Rendell L, Osborne RW, Wursig B. 2004. Culture and conservation ofnon humans with reference to whales and dolphins: review and new directions. Biological Conservation 120:427-437.

Whiten A. 2001. The evolutionary roots of culture. The British Academy Review July December 57-60.

Whiten A, Ham R. 1992. On the nature and evolution of imitation in the animal kingdom: reappraisal of a century of research. Advances in the Study ofBehavior 21 :239-283.

Whiten A, Goodall J, McGrew WC, Nishida T, Reynolds V, Sugiyama Y, Tutin CEG,Wrangham RW, Boesch C. 1999. Cultures in chimpanzees. Nature 399:682-685.

Whiten A, Goodall J, McGrew WC, Nishida T, Reynolds V, Sugiyama Y, Tutin CEG, Wrangham RW, Boesch C. 2001. Charting cultural variation in chimpanzees. Behaviour 138:1481-1516.

Wrangham RW. 1977. Feeding behaviour of chimpanzees in Gembe National Park, Tanzania. In: Clutton-Brock TH, editor. Primate Ecology: Studies offeeding and ranging behaviour in lemurs, monkeys and apes, London, New York: Academic Press p. 503-538.

Wrangham RW. 1980. An ecological model of female-bonded primate groups. Behaviour 75:262-299. 78

Wrangham RW, Conklin NL, Etot G, Obua J, Hunt KD, Hauser MD, Clark AP. 1993. The value of figs to chimpanzees. International Journal ofPrimatology 14:243-256.

Yamakoshi G. 1998. Dietary responses to fruit scarcity of wild chimpanzees at Bossou, Guinea: possible implications for ecological importance for tool use. American Journal of Physical Anthropology 106:283-295.